Menu
▾
▴
kaldi-users — Kaldi Users
You can subscribe to this list here.
| 2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(2) |
Aug
(2) |
Sep
(1) |
Oct
(1) |
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2012 |
Jan
|
Feb
|
Mar
(8) |
Apr
(4) |
May
(2) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2013 |
Jan
|
Feb
(2) |
Mar
(2) |
Apr
(7) |
May
(31) |
Jun
(40) |
Jul
(65) |
Aug
(37) |
Sep
(12) |
Oct
(57) |
Nov
(15) |
Dec
(35) |
| 2014 |
Jan
(3) |
Feb
(30) |
Mar
(57) |
Apr
(26) |
May
(49) |
Jun
(26) |
Jul
(63) |
Aug
(33) |
Sep
(20) |
Oct
(153) |
Nov
(62) |
Dec
(20) |
| 2015 |
Jan
(6) |
Feb
(21) |
Mar
(42) |
Apr
(33) |
May
(76) |
Jun
(102) |
Jul
(39) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-18 07:03:46
|
Thank you for your prompt and accurate replies. The script was indeed finding an incompatible HCLG.fst from a previous run and would not update it:
if [ -s $dir/dengraph/HCLG.fst ]; then
echo Not creating denominator graph $dir/dengraph/HCLG.fst since it already exists.
else
scripts/mkgraph.sh $dir/lang $alidir $dir/dengraph || exit 1;
fi
I removed the old fst and so it was rebuilt … Things are working fine now.
nassos
On Jun 17, 2013, at 11:26 PM, Mailing list used for User Communication and Updates wrote:
> Is not this a mismatch between feature and model dimensions?
> Petr
>
>
>
> 2013/6/17 Mailing list used for User Communication and Updates <kal...@li...>
> It looks to me like your HCLG.fst may be out of date with respect to your model-- perhaps you changed your model and the HCLG.fst did not get rebuilt?
> Dan
>
>
>
> On Mon, Jun 17, 2013 at 1:51 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote:
> Hi all,
>
> I'm trying to run the s3 recipe for WSJ and I'm running into a problem I was wondering whether you could shed any light upon:
>
> The script run.sh works fine up to steps/align_lda_mllt.sh. However, when I am running:
>
> steps/make_denlats_lda_etc.sh --num-jobs 4 --cmd "$train_cmd" \
> data/train_si84 data/lang exp/tri2b_ali_si84 exp/tri2b_denlats_si84
>
> I get a KALDI_ASSERT error. I updated to the most recent version of the trunk and retried but with no effect.
>
> More specifically, I get the following output in one of the log files (I'm running the recipe on a cluster and I'm submitting to a queue using 20 jobs):
>
> =====================================
> >> cat wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
>
> Running on ro
> Started at Tue Jun 18 05:27:00 EEST 2013
> gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 --max-active=5000 --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst 'ark:apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- | splice-feats ark:- ark:- | transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- |' 'ark:|gzip -c >exp/tri2b_denlats_si84/lat.24.gz'
> splice-feats ark:- ark:-
> apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:-
> transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:-
> KALDI_ASSERT: at gmm-latgen-faster:TransitionIdToPdf:hmm/transition-model.h:309, failed: static_cast<size_t>(trans_id) < id2state_.size()
> Stack trace is:
> kaldi::KaldiGetStackTrace()
> kaldi::KaldiAssertFailure_(char const*, char const*, int, char const*)
> kaldi::TransitionModel::TransitionIdToPdf(int) const
> kaldi::DecodableAmDiagGmmScaled::LogLikelihood(int, int)
> kaldi::LatticeFasterDecoder::ProcessEmitting(kaldi::DecodableInterface*, int)
> kaldi::LatticeFasterDecoder::Decode(kaldi::DecodableInterface*)
> kaldi::DecodeUtteranceLatticeFaster(kaldi::LatticeFasterDecoder&, kaldi::DecodableInterface&, fst::SymbolTable const*, std::string, double, bool, bool, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, kaldi::TableWriter<kaldi::CompactLatticeHolder>*, kaldi::TableWriter<kaldi::LatticeHolder>*, double*)
> gmm-latgen-faster(main+0xc3b) [0x58dad6]
> /lib64/libc.so.6(__libc_start_main+0xe6) [0x2ba2f7d9cc16]
> gmm-latgen-faster() [0x58cd11]
> /rmt/programs/gridengine_new/default/spool/ro/job_scripts/10778: line 6: 26822 Aborted (core dumped) ( gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 --max-active=5000 --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst "ark:apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- | splice-feats ark:- ark:- | transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c >exp/tri2b_denlats_si84/lat.24.gz" ) 2>> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
>
> =====================================
>
> I've started looking into the code in further detail but I guess debugging in this way will take a while since I have very little experience with kaldi. So, any ideas or suggestions will be greatly appreciated.
>
> Thank you,
> nassos
>
>
>
>
> PS: The decode_den.24.sh script:
>
> =====================================
>
> #!/bin/bash
> cd /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3
> . path.sh
> echo Running on `hostname` >/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
> echo Started at `date` >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
> ( gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 --max-active=5000 --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst "ark:apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- | splice-feats ark:- ark:- | transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c >exp/tri2b_denlats_si84/lat.24.gz" ) 2>>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
> ret=$?
> echo >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
> echo Finished at `date` >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log
> exit $ret
> ## submitted with:
> # qsub -S /bin/bash -sync y -j y -o /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log -l mem_free=700M /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/decode_den.24.sh >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/queue.log 2>&1
> =====================================
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev
> _______________________________________________
> Kaldi-users mailing list
> Kal...@li...
> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev
> _______________________________________________
> Kaldi-users mailing list
> Kal...@li...
> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by Windows:
>
> Build for Windows Store.
>
> http://p.sf.net/sfu/windows-dev2dev_______________________________________________
> Kaldi-users mailing list
> Kal...@li...
> https://lists.sourceforge.net/lists/listinfo/kaldi-users
|
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-17 20:26:51
|
Is not this a mismatch between feature and model dimensions? Petr 2013/6/17 Mailing list used for User Communication and Updates < kal...@li...> > It looks to me like your HCLG.fst may be out of date with respect to your > model-- perhaps you changed your model and the HCLG.fst did not get rebuilt? > Dan > > > > On Mon, Jun 17, 2013 at 1:51 PM, Mailing list used for User Communication > and Updates <kal...@li...> wrote: > >> Hi all, >> >> I'm trying to run the s3 recipe for WSJ and I'm running into a problem I >> was wondering whether you could shed any light upon: >> >> The script run.sh works fine up to steps/align_lda_mllt.sh. However, when >> I am running: >> >> steps/make_denlats_lda_etc.sh --num-jobs 4 --cmd "$train_cmd" \ >> data/train_si84 data/lang exp/tri2b_ali_si84 exp/tri2b_denlats_si84 >> >> I get a KALDI_ASSERT error. I updated to the most recent version of the >> trunk and retried but with no effect. >> >> More specifically, I get the following output in one of the log files >> (I'm running the recipe on a cluster and I'm submitting to a queue using 20 >> jobs): >> >> ===================================== >> >> cat wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> >> Running on ro >> Started at Tue Jun 18 05:27:00 EEST 2013 >> gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 >> --max-mem=20000000 --max-active=5000 >> --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl >> exp/tri2b_denlats_si84/dengraph/HCLG.fst 'ark:apply-cmvn --norm-vars=false >> --utt2spk=ark:data/train_si84/split20/24/utt2spk >> ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp >> ark:- | splice-feats ark:- ark:- | transform-feats >> exp/tri2b_ali_si84/final.mat ark:- ark:- |' 'ark:|gzip -c >> >exp/tri2b_denlats_si84/lat.24.gz' >> splice-feats ark:- ark:- >> apply-cmvn --norm-vars=false >> --utt2spk=ark:data/train_si84/split20/24/utt2spk >> ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp >> ark:- >> transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- >> KALDI_ASSERT: at >> gmm-latgen-faster:TransitionIdToPdf:hmm/transition-model.h:309, failed: >> static_cast<size_t>(trans_id) < id2state_.size() >> Stack trace is: >> kaldi::KaldiGetStackTrace() >> kaldi::KaldiAssertFailure_(char const*, char const*, int, char const*) >> kaldi::TransitionModel::TransitionIdToPdf(int) const >> kaldi::DecodableAmDiagGmmScaled::LogLikelihood(int, int) >> kaldi::LatticeFasterDecoder::ProcessEmitting(kaldi::DecodableInterface*, >> int) >> kaldi::LatticeFasterDecoder::Decode(kaldi::DecodableInterface*) >> kaldi::DecodeUtteranceLatticeFaster(kaldi::LatticeFasterDecoder&, >> kaldi::DecodableInterface&, fst::SymbolTable const*, std::string, double, >> bool, bool, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, >> kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, >> kaldi::TableWriter<kaldi::CompactLatticeHolder>*, >> kaldi::TableWriter<kaldi::LatticeHolder>*, double*) >> gmm-latgen-faster(main+0xc3b) [0x58dad6] >> /lib64/libc.so.6(__libc_start_main+0xe6) [0x2ba2f7d9cc16] >> gmm-latgen-faster() [0x58cd11] >> /rmt/programs/gridengine_new/default/spool/ro/job_scripts/10778: line 6: >> 26822 Aborted (core dumped) ( gmm-latgen-faster --beam=13.0 >> --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 >> --max-active=5000 --word-symbol-table=data/lang/words.txt >> exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst >> "ark:apply-cmvn --norm-vars=false >> --utt2spk=ark:data/train_si84/split20/24/utt2spk >> ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp >> ark:- | splice-feats ark:- ark:- | transform-feats >> exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c >> >exp/tri2b_denlats_si84/lat.24.gz" ) 2>> >> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> >> >> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> >> ===================================== >> >> I've started looking into the code in further detail but I guess >> debugging in this way will take a while since I have very little experience >> with kaldi. So, any ideas or suggestions will be greatly appreciated. >> >> Thank you, >> nassos >> >> >> >> >> PS: The decode_den.24.sh script: >> >> ===================================== >> >> #!/bin/bash >> cd /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3 >> . path.sh >> echo Running on `hostname` >> >/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> echo Started at `date` >> >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> ( gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 >> --max-mem=20000000 --max-active=5000 >> --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl >> exp/tri2b_denlats_si84/dengraph/HCLG.fst "ark:apply-cmvn --norm-vars=false >> --utt2spk=ark:data/train_si84/split20/24/utt2spk >> ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp >> ark:- | splice-feats ark:- ark:- | transform-feats >> exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c >> >exp/tri2b_denlats_si84/lat.24.gz" ) >> 2>>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> ret=$? >> echo >> >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> echo Finished at `date` >> >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> exit $ret >> ## submitted with: >> # qsub -S /bin/bash -sync y -j y -o >> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> -l mem_free=700M >> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/ >> decode_den.24.sh>>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/queue.log >> 2>&1 >> ===================================== >> >> >> >> ------------------------------------------------------------------------------ >> This SF.net email is sponsored by Windows: >> >> Build for Windows Store. >> >> http://p.sf.net/sfu/windows-dev2dev >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > > > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by Windows: > > Build for Windows Store. > > http://p.sf.net/sfu/windows-dev2dev > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-17 19:54:04
|
It looks to me like your HCLG.fst may be out of date with respect to your model-- perhaps you changed your model and the HCLG.fst did not get rebuilt? Dan On Mon, Jun 17, 2013 at 1:51 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Hi all, > > I'm trying to run the s3 recipe for WSJ and I'm running into a problem I > was wondering whether you could shed any light upon: > > The script run.sh works fine up to steps/align_lda_mllt.sh. However, when > I am running: > > steps/make_denlats_lda_etc.sh --num-jobs 4 --cmd "$train_cmd" \ > data/train_si84 data/lang exp/tri2b_ali_si84 exp/tri2b_denlats_si84 > > I get a KALDI_ASSERT error. I updated to the most recent version of the > trunk and retried but with no effect. > > More specifically, I get the following output in one of the log files (I'm > running the recipe on a cluster and I'm submitting to a queue using 20 > jobs): > > ===================================== > >> cat wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > > Running on ro > Started at Tue Jun 18 05:27:00 EEST 2013 > gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 > --max-mem=20000000 --max-active=5000 > --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl > exp/tri2b_denlats_si84/dengraph/HCLG.fst 'ark:apply-cmvn --norm-vars=false > --utt2spk=ark:data/train_si84/split20/24/utt2spk > ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp > ark:- | splice-feats ark:- ark:- | transform-feats > exp/tri2b_ali_si84/final.mat ark:- ark:- |' 'ark:|gzip -c > >exp/tri2b_denlats_si84/lat.24.gz' > splice-feats ark:- ark:- > apply-cmvn --norm-vars=false > --utt2spk=ark:data/train_si84/split20/24/utt2spk > ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp > ark:- > transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- > KALDI_ASSERT: at > gmm-latgen-faster:TransitionIdToPdf:hmm/transition-model.h:309, failed: > static_cast<size_t>(trans_id) < id2state_.size() > Stack trace is: > kaldi::KaldiGetStackTrace() > kaldi::KaldiAssertFailure_(char const*, char const*, int, char const*) > kaldi::TransitionModel::TransitionIdToPdf(int) const > kaldi::DecodableAmDiagGmmScaled::LogLikelihood(int, int) > kaldi::LatticeFasterDecoder::ProcessEmitting(kaldi::DecodableInterface*, > int) > kaldi::LatticeFasterDecoder::Decode(kaldi::DecodableInterface*) > kaldi::DecodeUtteranceLatticeFaster(kaldi::LatticeFasterDecoder&, > kaldi::DecodableInterface&, fst::SymbolTable const*, std::string, double, > bool, bool, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, > kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, > kaldi::TableWriter<kaldi::CompactLatticeHolder>*, > kaldi::TableWriter<kaldi::LatticeHolder>*, double*) > gmm-latgen-faster(main+0xc3b) [0x58dad6] > /lib64/libc.so.6(__libc_start_main+0xe6) [0x2ba2f7d9cc16] > gmm-latgen-faster() [0x58cd11] > /rmt/programs/gridengine_new/default/spool/ro/job_scripts/10778: line 6: > 26822 Aborted (core dumped) ( gmm-latgen-faster --beam=13.0 > --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 > --max-active=5000 --word-symbol-table=data/lang/words.txt > exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst > "ark:apply-cmvn --norm-vars=false > --utt2spk=ark:data/train_si84/split20/24/utt2spk > ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp > ark:- | splice-feats ark:- ark:- | transform-feats > exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c > >exp/tri2b_denlats_si84/lat.24.gz" ) 2>> > /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > >> > /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > > ===================================== > > I've started looking into the code in further detail but I guess debugging > in this way will take a while since I have very little experience with > kaldi. So, any ideas or suggestions will be greatly appreciated. > > Thank you, > nassos > > > > > PS: The decode_den.24.sh script: > > ===================================== > > #!/bin/bash > cd /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3 > . path.sh > echo Running on `hostname` > >/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > echo Started at `date` > >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > ( gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 > --max-mem=20000000 --max-active=5000 > --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl > exp/tri2b_denlats_si84/dengraph/HCLG.fst "ark:apply-cmvn --norm-vars=false > --utt2spk=ark:data/train_si84/split20/24/utt2spk > ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp > ark:- | splice-feats ark:- ark:- | transform-feats > exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c > >exp/tri2b_denlats_si84/lat.24.gz" ) > 2>>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > ret=$? > echo > >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > echo Finished at `date` > >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > exit $ret > ## submitted with: > # qsub -S /bin/bash -sync y -j y -o > /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log > -l mem_free=700M > /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/ > decode_den.24.sh>>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/queue.log > 2>&1 > ===================================== > > > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by Windows: > > Build for Windows Store. > > http://p.sf.net/sfu/windows-dev2dev > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-17 17:51:56
|
Hi all, I'm trying to run the s3 recipe for WSJ and I'm running into a problem I was wondering whether you could shed any light upon: The script run.sh works fine up to steps/align_lda_mllt.sh. However, when I am running: steps/make_denlats_lda_etc.sh --num-jobs 4 --cmd "$train_cmd" \ data/train_si84 data/lang exp/tri2b_ali_si84 exp/tri2b_denlats_si84 I get a KALDI_ASSERT error. I updated to the most recent version of the trunk and retried but with no effect. More specifically, I get the following output in one of the log files (I'm running the recipe on a cluster and I'm submitting to a queue using 20 jobs): ===================================== >> cat wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log Running on ro Started at Tue Jun 18 05:27:00 EEST 2013 gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 --max-active=5000 --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst 'ark:apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- | splice-feats ark:- ark:- | transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- |' 'ark:|gzip -c >exp/tri2b_denlats_si84/lat.24.gz' splice-feats ark:- ark:- apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- KALDI_ASSERT: at gmm-latgen-faster:TransitionIdToPdf:hmm/transition-model.h:309, failed: static_cast<size_t>(trans_id) < id2state_.size() Stack trace is: kaldi::KaldiGetStackTrace() kaldi::KaldiAssertFailure_(char const*, char const*, int, char const*) kaldi::TransitionModel::TransitionIdToPdf(int) const kaldi::DecodableAmDiagGmmScaled::LogLikelihood(int, int) kaldi::LatticeFasterDecoder::ProcessEmitting(kaldi::DecodableInterface*, int) kaldi::LatticeFasterDecoder::Decode(kaldi::DecodableInterface*) kaldi::DecodeUtteranceLatticeFaster(kaldi::LatticeFasterDecoder&, kaldi::DecodableInterface&, fst::SymbolTable const*, std::string, double, bool, bool, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, kaldi::TableWriter<kaldi::BasicVectorHolder<int> >*, kaldi::TableWriter<kaldi::CompactLatticeHolder>*, kaldi::TableWriter<kaldi::LatticeHolder>*, double*) gmm-latgen-faster(main+0xc3b) [0x58dad6] /lib64/libc.so.6(__libc_start_main+0xe6) [0x2ba2f7d9cc16] gmm-latgen-faster() [0x58cd11] /rmt/programs/gridengine_new/default/spool/ro/job_scripts/10778: line 6: 26822 Aborted (core dumped) ( gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 --max-active=5000 --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst "ark:apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- | splice-feats ark:- ark:- | transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c >exp/tri2b_denlats_si84/lat.24.gz" ) 2>> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >> /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log ===================================== I've started looking into the code in further detail but I guess debugging in this way will take a while since I have very little experience with kaldi. So, any ideas or suggestions will be greatly appreciated. Thank you, nassos PS: The decode_den.24.sh script: ===================================== #!/bin/bash cd /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3 . path.sh echo Running on `hostname` >/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log echo Started at `date` >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log ( gmm-latgen-faster --beam=13.0 --lattice-beam=7.0 --acoustic-scale=0.1 --max-mem=20000000 --max-active=5000 --word-symbol-table=data/lang/words.txt exp/tri2b_ali_si84/final.mdl exp/tri2b_denlats_si84/dengraph/HCLG.fst "ark:apply-cmvn --norm-vars=false --utt2spk=ark:data/train_si84/split20/24/utt2spk ark:exp/tri2b_ali_si84/24.cmvn scp:data/train_si84/split20/24/feats.scp ark:- | splice-feats ark:- ark:- | transform-feats exp/tri2b_ali_si84/final.mat ark:- ark:- |" "ark:|gzip -c >exp/tri2b_denlats_si84/lat.24.gz" ) 2>>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log ret=$? echo >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log echo Finished at `date` >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log exit $ret ## submitted with: # qsub -S /bin/bash -sync y -j y -o /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/decode_den.24.log -l mem_free=700M /rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/decode_den.24.sh >>/rmt/work/audio_asr/kaldi/kaldi-trunk/egs/wsj/s3/exp/tri2b_denlats_si84/q/queue.log 2>&1 ===================================== |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 19:33:14
|
Yes, some small variation is expected Thanks for letting us know! Dan On Tue, Jun 4, 2013 at 3:10 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Yes, I can confirm the fix. Here are my results. They are in the same > ballpark, so I'm happy enough. > > %WER 37.86 [ 3117 / 8234, 267 ins, 579 del, 2271 sub ] > exp/mono0a_nj4/decode_tgpr_dev93/wer_10 > > %WER 26.78 [ 1511 / 5643, 163 ins, 206 del, 1142 sub ] > exp/mono0a_nj4/decode_tgpr_eval92/wer_10 > > Thank you so much. > > Hao > > On Tue, Jun 4, 2013 at 12:15 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > Thanks [arnab, I assume]. > > I just found the problem myself also. > > Sorry. > > > > Dan > > > > > > > > On Tue, Jun 4, 2013 at 1:08 PM, Mailing list used for User Communication > and > > Updates <kal...@li...> wrote: > >> > >> Found the problem! It's in local/score.sh -- in rev 2443 you added an > >> option for using word insertion penalty before which the lattice > >> acoustic scores get scaled by lm weight. But the lattice-best-path is > >> also using a scaling, and so the scores are being scaled twice. The > >> fix is in the repository. But I haven't checked if the word insertion > >> penalty has been added to other scoring scripts as well. > >> > >> Hao, please rerun and you should get something around 34.76% for dev93. > >> > >> On Tue, Jun 4, 2013 at 5:48 PM, Mailing list used for User > >> Communication and Updates <kal...@li...> wrote: > >> > BTW, I suspect local/make_lexicon_fst.pl > >> > I made a change which I thought was a simple bug fix that would not > >> > affect > >> > anything. > >> > Dan > >> > > >> > > >> > > >> > On Tue, Jun 4, 2013 at 12:33 PM, Mailing list used for User > >> > Communication > >> > and Updates <kal...@li...> wrote: > >> >> > >> >> There seems to be something broken with the current WSJ setup. I > reran > >> >> the setup today and I am getting similar results as Hao. For example, > >> >> an older setup gets 34.98% on dev93, while the system trained today > >> >> gets 47.79%. The problem is not features, since the older model with > >> >> the new features is 34.76%. The problem is also not the dict/lm since > >> >> the old model with new dict/lm and old features gets 34.98%. It is > >> >> unlikely to be the new model, since the log-likelihood numbers are in > >> >> the same ballpark. My current best guess is that some setting in the > >> >> decode/scoring scripts has changed, which is causing the problem. I > am > >> >> looking into this in more detail. > >> >> -Arnab > >> >> > >> >> On Tue, Jun 4, 2013 at 8:02 AM, Mailing list used for User > >> >> Communication and Updates <kal...@li...> wrote: > >> >> > Here are the log for the last update with different number of > jobs. I > >> >> > don't see much of a difference. > >> >> > > >> >> > In terms of likelihood shown below, they are pretty consistent. > >> >> > > >> >> > nj=2 -109.265 -> -95.5657 > >> >> > nj=4 -109.265 -> -95.5599 > >> >> > nj=8 -109.265 -> -95.556 > >> >> > > >> >> > Hao > >> >> > > >> >> > On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User > >> >> > Communication and Updates <kal...@li...> > wrote: > >> >> >> OK, I have an idea what might have happened. It's possible that > >> >> >> because you > >> >> >> are using such a small number of parallel jobs (2), there is a > >> >> >> problem > >> >> >> due > >> >> >> to numerical roundoff. This should not be the case because we > >> >> >> accumulate in > >> >> >> double, but to see whether this is a factor, could you please run > >> >> >> the > >> >> >> training with more jobs? E.g. 4 or 8? See if the likelihoods > >> >> >> change. > >> >> >> Also, > >> >> >> show me the full log from the last update, both before and after > you > >> >> >> make > >> >> >> the change. > >> >> >> Dan > >> >> >> > >> >> >> > >> >> >> > >> >> >> On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User > >> >> >> Communication and > >> >> >> Updates <kal...@li...> wrote: > >> >> >>> > >> >> >>> They are almost identical. > >> >> >>> > >> >> >>> 011c0201 [ > >> >> >>> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 > >> >> >>> -5.20842 > >> >> >>> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 > >> >> >>> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 > >> >> >>> -12.17628 > >> >> >>> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 > >> >> >>> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 > >> >> >>> -3.351704 > >> >> >>> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 > >> >> >>> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 > -10.907 > >> >> >>> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > >> >> >>> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 > >> >> >>> -9.91317 > >> >> >>> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 > >> >> >>> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 > >> >> >>> -8.361964 > >> >> >>> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > >> >> >>> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 > >> >> >>> -3.151051 > >> >> >>> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 > >> >> >>> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 > >> >> >>> -1.814186 > >> >> >>> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 > >> >> >>> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 > >> >> >>> -10.80855 > >> >> >>> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 > >> >> >>> > >> >> >>> Hao > >> >> >>> > >> >> >>> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User > >> >> >>> Communication and Updates <kal...@li...> > >> >> >>> wrote: > >> >> >>> > OK, everything there looks pretty normal. Let's check the > >> >> >>> > features. > >> >> >>> > Try > >> >> >>> > doing > >> >> >>> > > >> >> >>> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | > >> >> >>> > head > >> >> >>> > copy-feats scp:data/train_si284/feats.scp ark,t:- > >> >> >>> > 011c0201 [ > >> >> >>> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 > >> >> >>> > -5.20842 > >> >> >>> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 > >> >> >>> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 > >> >> >>> > -12.17628 > >> >> >>> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 > >> >> >>> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 > >> >> >>> > -3.351702 > >> >> >>> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 > >> >> >>> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 > >> >> >>> > -10.907 > >> >> >>> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > >> >> >>> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 > >> >> >>> > -9.913167 > >> >> >>> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 > >> >> >>> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 > >> >> >>> > -8.361964 > >> >> >>> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > >> >> >>> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 > >> >> >>> > -3.151051 > >> >> >>> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 > >> >> >>> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 > >> >> >>> > -1.814186 > >> >> >>> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 > >> >> >>> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 > >> >> >>> > -10.80855 > >> >> >>> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 > >> >> >>> > > >> >> >>> > and show me the otuput > >> >> >>> > Dan > >> >> >>> > > >> >> >>> > > >> >> >>> > > >> >> >>> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User > >> >> >>> > Communication > >> >> >>> > and > >> >> >>> > Updates <kal...@li...> wrote: > >> >> >>> >> > >> >> >>> >> Here is the result for > >> >> >>> >> > >> >> >>> >> ls -l data/lang data/lang/phones exp/mono0a > exp/mono0a/log > >> >> >>> >> exp/mono0a/decode/log > >> >> >>> >> > >> >> >>> >> Hao > >> >> >>> >> > >> >> >>> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User > >> >> >>> >> Communication and Updates <kal...@li...> > >> >> >>> >> wrote: > >> >> >>> >> > My corresponding numbers are -108.535 -> -95.116, and the > >> >> >>> >> > difference > >> >> >>> >> > is > >> >> >>> >> > about the same. [It's strange that the absolute numbers > >> >> >>> >> > differ > >> >> >>> >> > slightly, > >> >> >>> >> > though.] > >> >> >>> >> > Could you do > >> >> >>> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >> >> >>> >> > exp/mono0a/decode/log > >> >> >>> >> > and send the output? > >> >> >>> >> > > >> >> >>> >> > Dan > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User > >> >> >>> >> > Communication > >> >> >>> >> > and > >> >> >>> >> > Updates <kal...@li...> wrote: > >> >> >>> >> >> > >> >> >>> >> >> Copied from exp/mono0a/log/update.39.log > >> >> >>> >> >> > >> >> >>> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 > >> >> >>> >> >> stats, > >> >> >>> >> >> total > >> >> >>> >> >> count 849568, avg like/frame -95.5657 > >> >> >>> >> >> > >> >> >>> >> >> Hao > >> >> >>> >> >> > >> >> >>> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User > >> >> >>> >> >> Communication and Updates < > kal...@li...> > >> >> >>> >> >> wrote: > >> >> >>> >> >> > OK, that is quite weird. What is the avg like/frame on > the > >> >> >>> >> >> > final > >> >> >>> >> >> > update.??.log? > >> >> >>> >> >> > Dan > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for > User > >> >> >>> >> >> > Communication > >> >> >>> >> >> > and > >> >> >>> >> >> > Updates <kal...@li...> wrote: > >> >> >>> >> >> >> > >> >> >>> >> >> >> Unfortunately those are absolute differences. Here is > what > >> >> >>> >> >> >> I > >> >> >>> >> >> >> get > >> >> >>> >> >> >> for > >> >> >>> >> >> >> the > >> >> >>> >> >> >> mono0a. > >> >> >>> >> >> >> > >> >> >>> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 > sub ] > >> >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 > >> >> >>> >> >> >> > >> >> >>> >> >> >> And below is the reported one in RESULTS. > >> >> >>> >> >> >> > >> >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ > 1441 > >> >> >>> >> >> >> / > >> >> >>> >> >> >> 5643, > >> >> >>> >> >> >> 149 > >> >> >>> >> >> >> ins, 191 del, 1101 sub ] > >> >> >>> >> >> >> > >> >> >>> >> >> >> I will try to checkout the same version as the RESULTS > >> >> >>> >> >> >> file > >> >> >>> >> >> >> and > >> >> >>> >> >> >> run > >> >> >>> >> >> >> the experiments again. > >> >> >>> >> >> >> > >> >> >>> >> >> >> Here is the total data count from > >> >> >>> >> >> >> exp/mono0a/log/update.0.log. > >> >> >>> >> >> >> > >> >> >>> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed > 2 > >> >> >>> >> >> >> stats, > >> >> >>> >> >> >> total > >> >> >>> >> >> >> count 849568, avg like/frame -109.265 > >> >> >>> >> >> >> > >> >> >>> >> >> >> Thanks for your help. > >> >> >>> >> >> >> > >> >> >>> >> >> >> Hao > >> >> >>> >> >> >> > >> >> >>> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for > User > >> >> >>> >> >> >> Communication and Updates > >> >> >>> >> >> >> <kal...@li...> > >> >> >>> >> >> >> wrote: > >> >> >>> >> >> >> > Assuming those are relative numbers, they're probably > >> >> >>> >> >> >> > within > >> >> >>> >> >> >> > the > >> >> >>> >> >> >> > margin > >> >> >>> >> >> >> > of > >> >> >>> >> >> >> > error. See if it's the same on other test sets. > >> >> >>> >> >> >> > Because > >> >> >>> >> >> >> > various > >> >> >>> >> >> >> > algorithms > >> >> >>> >> >> >> > call rand(), and different machines implement this > >> >> >>> >> >> >> > differently, > >> >> >>> >> >> >> > results > >> >> >>> >> >> >> > aren't fully reproducible. > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > But to check that it's not some code change that hurts > >> >> >>> >> >> >> > the > >> >> >>> >> >> >> > results, > >> >> >>> >> >> >> > you > >> >> >>> >> >> >> > could try checking out a copy with the same revision > >> >> >>> >> >> >> > number > >> >> >>> >> >> >> > as > >> >> >>> >> >> >> > the > >> >> >>> >> >> >> > RESULTS > >> >> >>> >> >> >> > file, and running again and seeing what results you > get. > >> >> >>> >> >> >> > If > >> >> >>> >> >> >> > there > >> >> >>> >> >> >> > is > >> >> >>> >> >> >> > a > >> >> >>> >> >> >> > difference, I'd like to know. > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > It's also possible some files were missing in your WSJ > >> >> >>> >> >> >> > distribution-- > >> >> >>> >> >> >> > let me > >> >> >>> >> >> >> > know the data count reported in one the > */log/update.log > >> >> >>> >> >> >> > files > >> >> >>> >> >> >> > and > >> >> >>> >> >> >> > I'll > >> >> >>> >> >> >> > compare with a local copy. > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > Dan > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for > >> >> >>> >> >> >> > User > >> >> >>> >> >> >> > Communication > >> >> >>> >> >> >> > and > >> >> >>> >> >> >> > Updates <kal...@li...> wrote: > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> Hi, > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to > >> >> >>> >> >> >> >> reproduce > >> >> >>> >> >> >> >> the > >> >> >>> >> >> >> >> results stated in the RESULTS file. Are we supposed > to > >> >> >>> >> >> >> >> get > >> >> >>> >> >> >> >> exactly > >> >> >>> >> >> >> >> the > >> >> >>> >> >> >> >> same numbers with the (hyper-)parameters given in the > >> >> >>> >> >> >> >> recipe? > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> Just to give a sense of how far I'm off. My monophone > >> >> >>> >> >> >> >> models > >> >> >>> >> >> >> >> perform > >> >> >>> >> >> >> >> 10% > >> >> >>> >> >> >> >> worse than the reported results on eval92 and the > >> >> >>> >> >> >> >> resulting > >> >> >>> >> >> >> >> triphone > >> >> >>> >> >> >> >> models perform 4% worse. > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> Thanks, > >> >> >>> >> >> >> >> Hao > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > ------------------------------------------------------------------------------ > >> >> >>> >> >> >> >> How ServiceNow helps IT people transform IT > >> >> >>> >> >> >> >> departments: > >> >> >>> >> >> >> >> 1. A cloud service to automate IT design, transition > >> >> >>> >> >> >> >> and > >> >> >>> >> >> >> >> operations > >> >> >>> >> >> >> >> 2. Dashboards that offer high-level views of > enterprise > >> >> >>> >> >> >> >> services > >> >> >>> >> >> >> >> 3. A single system of record for all IT processes > >> >> >>> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> >> >> >> _______________________________________________ > >> >> >>> >> >> >> >> Kaldi-users mailing list > >> >> >>> >> >> >> >> Kal...@li... > >> >> >>> >> >> >> >> > >> >> >>> >> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > > ------------------------------------------------------------------------------ > >> >> >>> >> >> >> > How ServiceNow helps IT people transform IT > departments: > >> >> >>> >> >> >> > 1. A cloud service to automate IT design, transition > and > >> >> >>> >> >> >> > operations > >> >> >>> >> >> >> > 2. Dashboards that offer high-level views of > enterprise > >> >> >>> >> >> >> > services > >> >> >>> >> >> >> > 3. A single system of record for all IT processes > >> >> >>> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> >> >> > _______________________________________________ > >> >> >>> >> >> >> > Kaldi-users mailing list > >> >> >>> >> >> >> > Kal...@li... > >> >> >>> >> >> >> > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> >> >> > > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > ------------------------------------------------------------------------------ > >> >> >>> >> >> >> How ServiceNow helps IT people transform IT departments: > >> >> >>> >> >> >> 1. A cloud service to automate IT design, transition and > >> >> >>> >> >> >> operations > >> >> >>> >> >> >> 2. Dashboards that offer high-level views of enterprise > >> >> >>> >> >> >> services > >> >> >>> >> >> >> 3. A single system of record for all IT processes > >> >> >>> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> >> >> _______________________________________________ > >> >> >>> >> >> >> Kaldi-users mailing list > >> >> >>> >> >> >> Kal...@li... > >> >> >>> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > >> >> >>> >> >> > > ------------------------------------------------------------------------------ > >> >> >>> >> >> > How ServiceNow helps IT people transform IT departments: > >> >> >>> >> >> > 1. A cloud service to automate IT design, transition and > >> >> >>> >> >> > operations > >> >> >>> >> >> > 2. Dashboards that offer high-level views of enterprise > >> >> >>> >> >> > services > >> >> >>> >> >> > 3. A single system of record for all IT processes > >> >> >>> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> >> > _______________________________________________ > >> >> >>> >> >> > Kaldi-users mailing list > >> >> >>> >> >> > Kal...@li... > >> >> >>> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> >> > > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > ------------------------------------------------------------------------------ > >> >> >>> >> >> How ServiceNow helps IT people transform IT departments: > >> >> >>> >> >> 1. A cloud service to automate IT design, transition and > >> >> >>> >> >> operations > >> >> >>> >> >> 2. Dashboards that offer high-level views of enterprise > >> >> >>> >> >> services > >> >> >>> >> >> 3. A single system of record for all IT processes > >> >> >>> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> >> _______________________________________________ > >> >> >>> >> >> Kaldi-users mailing list > >> >> >>> >> >> Kal...@li... > >> >> >>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > >> >> >>> >> > > ------------------------------------------------------------------------------ > >> >> >>> >> > How ServiceNow helps IT people transform IT departments: > >> >> >>> >> > 1. A cloud service to automate IT design, transition and > >> >> >>> >> > operations > >> >> >>> >> > 2. Dashboards that offer high-level views of enterprise > >> >> >>> >> > services > >> >> >>> >> > 3. A single system of record for all IT processes > >> >> >>> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> > _______________________________________________ > >> >> >>> >> > Kaldi-users mailing list > >> >> >>> >> > Kal...@li... > >> >> >>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> > > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > ------------------------------------------------------------------------------ > >> >> >>> >> How ServiceNow helps IT people transform IT departments: > >> >> >>> >> 1. A cloud service to automate IT design, transition and > >> >> >>> >> operations > >> >> >>> >> 2. Dashboards that offer high-level views of enterprise > services > >> >> >>> >> 3. A single system of record for all IT processes > >> >> >>> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> >> _______________________________________________ > >> >> >>> >> Kaldi-users mailing list > >> >> >>> >> Kal...@li... > >> >> >>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> >> > >> >> >>> > > >> >> >>> > > >> >> >>> > > >> >> >>> > > >> >> >>> > > >> >> >>> > > ------------------------------------------------------------------------------ > >> >> >>> > How ServiceNow helps IT people transform IT departments: > >> >> >>> > 1. A cloud service to automate IT design, transition and > >> >> >>> > operations > >> >> >>> > 2. Dashboards that offer high-level views of enterprise > services > >> >> >>> > 3. A single system of record for all IT processes > >> >> >>> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> > _______________________________________________ > >> >> >>> > Kaldi-users mailing list > >> >> >>> > Kal...@li... > >> >> >>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >>> > > >> >> >>> > >> >> >>> > >> >> >>> > >> >> >>> > >> >> >>> > ------------------------------------------------------------------------------ > >> >> >>> How ServiceNow helps IT people transform IT departments: > >> >> >>> 1. A cloud service to automate IT design, transition and > operations > >> >> >>> 2. Dashboards that offer high-level views of enterprise services > >> >> >>> 3. A single system of record for all IT processes > >> >> >>> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >>> _______________________________________________ > >> >> >>> Kaldi-users mailing list > >> >> >>> Kal...@li... > >> >> >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > ------------------------------------------------------------------------------ > >> >> >> How ServiceNow helps IT people transform IT departments: > >> >> >> 1. A cloud service to automate IT design, transition and > operations > >> >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> >> 3. A single system of record for all IT processes > >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >> _______________________________________________ > >> >> >> Kaldi-users mailing list > >> >> >> Kal...@li... > >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >> > >> >> > > >> >> > > >> >> > > >> >> > > ------------------------------------------------------------------------------ > >> >> > How ServiceNow helps IT people transform IT departments: > >> >> > 1. A cloud service to automate IT design, transition and operations > >> >> > 2. Dashboards that offer high-level views of enterprise services > >> >> > 3. A single system of record for all IT processes > >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> > _______________________________________________ > >> >> > Kaldi-users mailing list > >> >> > Kal...@li... > >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ > >> >> How ServiceNow helps IT people transform IT departments: > >> >> 1. A cloud service to automate IT design, transition and operations > >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> 3. A single system of record for all IT processes > >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> _______________________________________________ > >> >> Kaldi-users mailing list > >> >> Kal...@li... > >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > > >> > > >> > > >> > > ------------------------------------------------------------------------------ > >> > How ServiceNow helps IT people transform IT departments: > >> > 1. A cloud service to automate IT design, transition and operations > >> > 2. Dashboards that offer high-level views of enterprise services > >> > 3. A single system of record for all IT processes > >> > http://p.sf.net/sfu/servicenow-d2d-j > >> > _______________________________________________ > >> > Kaldi-users mailing list > >> > Kal...@li... > >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 19:10:17
|
Yes, I can confirm the fix. Here are my results. They are in the same ballpark, so I'm happy enough. %WER 37.86 [ 3117 / 8234, 267 ins, 579 del, 2271 sub ] exp/mono0a_nj4/decode_tgpr_dev93/wer_10 %WER 26.78 [ 1511 / 5643, 163 ins, 206 del, 1142 sub ] exp/mono0a_nj4/decode_tgpr_eval92/wer_10 Thank you so much. Hao On Tue, Jun 4, 2013 at 12:15 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Thanks [arnab, I assume]. > I just found the problem myself also. > Sorry. > > Dan > > > > On Tue, Jun 4, 2013 at 1:08 PM, Mailing list used for User Communication and > Updates <kal...@li...> wrote: >> >> Found the problem! It's in local/score.sh -- in rev 2443 you added an >> option for using word insertion penalty before which the lattice >> acoustic scores get scaled by lm weight. But the lattice-best-path is >> also using a scaling, and so the scores are being scaled twice. The >> fix is in the repository. But I haven't checked if the word insertion >> penalty has been added to other scoring scripts as well. >> >> Hao, please rerun and you should get something around 34.76% for dev93. >> >> On Tue, Jun 4, 2013 at 5:48 PM, Mailing list used for User >> Communication and Updates <kal...@li...> wrote: >> > BTW, I suspect local/make_lexicon_fst.pl >> > I made a change which I thought was a simple bug fix that would not >> > affect >> > anything. >> > Dan >> > >> > >> > >> > On Tue, Jun 4, 2013 at 12:33 PM, Mailing list used for User >> > Communication >> > and Updates <kal...@li...> wrote: >> >> >> >> There seems to be something broken with the current WSJ setup. I reran >> >> the setup today and I am getting similar results as Hao. For example, >> >> an older setup gets 34.98% on dev93, while the system trained today >> >> gets 47.79%. The problem is not features, since the older model with >> >> the new features is 34.76%. The problem is also not the dict/lm since >> >> the old model with new dict/lm and old features gets 34.98%. It is >> >> unlikely to be the new model, since the log-likelihood numbers are in >> >> the same ballpark. My current best guess is that some setting in the >> >> decode/scoring scripts has changed, which is causing the problem. I am >> >> looking into this in more detail. >> >> -Arnab >> >> >> >> On Tue, Jun 4, 2013 at 8:02 AM, Mailing list used for User >> >> Communication and Updates <kal...@li...> wrote: >> >> > Here are the log for the last update with different number of jobs. I >> >> > don't see much of a difference. >> >> > >> >> > In terms of likelihood shown below, they are pretty consistent. >> >> > >> >> > nj=2 -109.265 -> -95.5657 >> >> > nj=4 -109.265 -> -95.5599 >> >> > nj=8 -109.265 -> -95.556 >> >> > >> >> > Hao >> >> > >> >> > On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User >> >> > Communication and Updates <kal...@li...> wrote: >> >> >> OK, I have an idea what might have happened. It's possible that >> >> >> because you >> >> >> are using such a small number of parallel jobs (2), there is a >> >> >> problem >> >> >> due >> >> >> to numerical roundoff. This should not be the case because we >> >> >> accumulate in >> >> >> double, but to see whether this is a factor, could you please run >> >> >> the >> >> >> training with more jobs? E.g. 4 or 8? See if the likelihoods >> >> >> change. >> >> >> Also, >> >> >> show me the full log from the last update, both before and after you >> >> >> make >> >> >> the change. >> >> >> Dan >> >> >> >> >> >> >> >> >> >> >> >> On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User >> >> >> Communication and >> >> >> Updates <kal...@li...> wrote: >> >> >>> >> >> >>> They are almost identical. >> >> >>> >> >> >>> 011c0201 [ >> >> >>> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 >> >> >>> -5.20842 >> >> >>> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 >> >> >>> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 >> >> >>> -12.17628 >> >> >>> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 >> >> >>> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 >> >> >>> -3.351704 >> >> >>> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 >> >> >>> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >> >> >>> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >> >> >>> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 >> >> >>> -9.91317 >> >> >>> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 >> >> >>> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 >> >> >>> -8.361964 >> >> >>> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >> >> >>> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 >> >> >>> -3.151051 >> >> >>> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 >> >> >>> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 >> >> >>> -1.814186 >> >> >>> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 >> >> >>> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 >> >> >>> -10.80855 >> >> >>> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 >> >> >>> >> >> >>> Hao >> >> >>> >> >> >>> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User >> >> >>> Communication and Updates <kal...@li...> >> >> >>> wrote: >> >> >>> > OK, everything there looks pretty normal. Let's check the >> >> >>> > features. >> >> >>> > Try >> >> >>> > doing >> >> >>> > >> >> >>> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | >> >> >>> > head >> >> >>> > copy-feats scp:data/train_si284/feats.scp ark,t:- >> >> >>> > 011c0201 [ >> >> >>> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 >> >> >>> > -5.20842 >> >> >>> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 >> >> >>> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 >> >> >>> > -12.17628 >> >> >>> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 >> >> >>> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 >> >> >>> > -3.351702 >> >> >>> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 >> >> >>> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 >> >> >>> > -10.907 >> >> >>> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >> >> >>> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 >> >> >>> > -9.913167 >> >> >>> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 >> >> >>> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 >> >> >>> > -8.361964 >> >> >>> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >> >> >>> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 >> >> >>> > -3.151051 >> >> >>> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 >> >> >>> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 >> >> >>> > -1.814186 >> >> >>> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 >> >> >>> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 >> >> >>> > -10.80855 >> >> >>> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 >> >> >>> > >> >> >>> > and show me the otuput >> >> >>> > Dan >> >> >>> > >> >> >>> > >> >> >>> > >> >> >>> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User >> >> >>> > Communication >> >> >>> > and >> >> >>> > Updates <kal...@li...> wrote: >> >> >>> >> >> >> >>> >> Here is the result for >> >> >>> >> >> >> >>> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> >> >>> >> exp/mono0a/decode/log >> >> >>> >> >> >> >>> >> Hao >> >> >>> >> >> >> >>> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User >> >> >>> >> Communication and Updates <kal...@li...> >> >> >>> >> wrote: >> >> >>> >> > My corresponding numbers are -108.535 -> -95.116, and the >> >> >>> >> > difference >> >> >>> >> > is >> >> >>> >> > about the same. [It's strange that the absolute numbers >> >> >>> >> > differ >> >> >>> >> > slightly, >> >> >>> >> > though.] >> >> >>> >> > Could you do >> >> >>> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> >> >>> >> > exp/mono0a/decode/log >> >> >>> >> > and send the output? >> >> >>> >> > >> >> >>> >> > Dan >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User >> >> >>> >> > Communication >> >> >>> >> > and >> >> >>> >> > Updates <kal...@li...> wrote: >> >> >>> >> >> >> >> >>> >> >> Copied from exp/mono0a/log/update.39.log >> >> >>> >> >> >> >> >>> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 >> >> >>> >> >> stats, >> >> >>> >> >> total >> >> >>> >> >> count 849568, avg like/frame -95.5657 >> >> >>> >> >> >> >> >>> >> >> Hao >> >> >>> >> >> >> >> >>> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User >> >> >>> >> >> Communication and Updates <kal...@li...> >> >> >>> >> >> wrote: >> >> >>> >> >> > OK, that is quite weird. What is the avg like/frame on the >> >> >>> >> >> > final >> >> >>> >> >> > update.??.log? >> >> >>> >> >> > Dan >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User >> >> >>> >> >> > Communication >> >> >>> >> >> > and >> >> >>> >> >> > Updates <kal...@li...> wrote: >> >> >>> >> >> >> >> >> >>> >> >> >> Unfortunately those are absolute differences. Here is what >> >> >>> >> >> >> I >> >> >>> >> >> >> get >> >> >>> >> >> >> for >> >> >>> >> >> >> the >> >> >>> >> >> >> mono0a. >> >> >>> >> >> >> >> >> >>> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] >> >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 >> >> >>> >> >> >> >> >> >>> >> >> >> And below is the reported one in RESULTS. >> >> >>> >> >> >> >> >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 >> >> >>> >> >> >> / >> >> >>> >> >> >> 5643, >> >> >>> >> >> >> 149 >> >> >>> >> >> >> ins, 191 del, 1101 sub ] >> >> >>> >> >> >> >> >> >>> >> >> >> I will try to checkout the same version as the RESULTS >> >> >>> >> >> >> file >> >> >>> >> >> >> and >> >> >>> >> >> >> run >> >> >>> >> >> >> the experiments again. >> >> >>> >> >> >> >> >> >>> >> >> >> Here is the total data count from >> >> >>> >> >> >> exp/mono0a/log/update.0.log. >> >> >>> >> >> >> >> >> >>> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 >> >> >>> >> >> >> stats, >> >> >>> >> >> >> total >> >> >>> >> >> >> count 849568, avg like/frame -109.265 >> >> >>> >> >> >> >> >> >>> >> >> >> Thanks for your help. >> >> >>> >> >> >> >> >> >>> >> >> >> Hao >> >> >>> >> >> >> >> >> >>> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User >> >> >>> >> >> >> Communication and Updates >> >> >>> >> >> >> <kal...@li...> >> >> >>> >> >> >> wrote: >> >> >>> >> >> >> > Assuming those are relative numbers, they're probably >> >> >>> >> >> >> > within >> >> >>> >> >> >> > the >> >> >>> >> >> >> > margin >> >> >>> >> >> >> > of >> >> >>> >> >> >> > error. See if it's the same on other test sets. >> >> >>> >> >> >> > Because >> >> >>> >> >> >> > various >> >> >>> >> >> >> > algorithms >> >> >>> >> >> >> > call rand(), and different machines implement this >> >> >>> >> >> >> > differently, >> >> >>> >> >> >> > results >> >> >>> >> >> >> > aren't fully reproducible. >> >> >>> >> >> >> > >> >> >>> >> >> >> > But to check that it's not some code change that hurts >> >> >>> >> >> >> > the >> >> >>> >> >> >> > results, >> >> >>> >> >> >> > you >> >> >>> >> >> >> > could try checking out a copy with the same revision >> >> >>> >> >> >> > number >> >> >>> >> >> >> > as >> >> >>> >> >> >> > the >> >> >>> >> >> >> > RESULTS >> >> >>> >> >> >> > file, and running again and seeing what results you get. >> >> >>> >> >> >> > If >> >> >>> >> >> >> > there >> >> >>> >> >> >> > is >> >> >>> >> >> >> > a >> >> >>> >> >> >> > difference, I'd like to know. >> >> >>> >> >> >> > >> >> >>> >> >> >> > It's also possible some files were missing in your WSJ >> >> >>> >> >> >> > distribution-- >> >> >>> >> >> >> > let me >> >> >>> >> >> >> > know the data count reported in one the */log/update.log >> >> >>> >> >> >> > files >> >> >>> >> >> >> > and >> >> >>> >> >> >> > I'll >> >> >>> >> >> >> > compare with a local copy. >> >> >>> >> >> >> > >> >> >>> >> >> >> > Dan >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for >> >> >>> >> >> >> > User >> >> >>> >> >> >> > Communication >> >> >>> >> >> >> > and >> >> >>> >> >> >> > Updates <kal...@li...> wrote: >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> Hi, >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to >> >> >>> >> >> >> >> reproduce >> >> >>> >> >> >> >> the >> >> >>> >> >> >> >> results stated in the RESULTS file. Are we supposed to >> >> >>> >> >> >> >> get >> >> >>> >> >> >> >> exactly >> >> >>> >> >> >> >> the >> >> >>> >> >> >> >> same numbers with the (hyper-)parameters given in the >> >> >>> >> >> >> >> recipe? >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> Just to give a sense of how far I'm off. My monophone >> >> >>> >> >> >> >> models >> >> >>> >> >> >> >> perform >> >> >>> >> >> >> >> 10% >> >> >>> >> >> >> >> worse than the reported results on eval92 and the >> >> >>> >> >> >> >> resulting >> >> >>> >> >> >> >> triphone >> >> >>> >> >> >> >> models perform 4% worse. >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> Thanks, >> >> >>> >> >> >> >> Hao >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> ------------------------------------------------------------------------------ >> >> >>> >> >> >> >> How ServiceNow helps IT people transform IT >> >> >>> >> >> >> >> departments: >> >> >>> >> >> >> >> 1. A cloud service to automate IT design, transition >> >> >>> >> >> >> >> and >> >> >>> >> >> >> >> operations >> >> >>> >> >> >> >> 2. Dashboards that offer high-level views of enterprise >> >> >>> >> >> >> >> services >> >> >>> >> >> >> >> 3. A single system of record for all IT processes >> >> >>> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> >> >> >> _______________________________________________ >> >> >>> >> >> >> >> Kaldi-users mailing list >> >> >>> >> >> >> >> Kal...@li... >> >> >>> >> >> >> >> >> >> >>> >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > >> >> >>> >> >> >> > ------------------------------------------------------------------------------ >> >> >>> >> >> >> > How ServiceNow helps IT people transform IT departments: >> >> >>> >> >> >> > 1. A cloud service to automate IT design, transition and >> >> >>> >> >> >> > operations >> >> >>> >> >> >> > 2. Dashboards that offer high-level views of enterprise >> >> >>> >> >> >> > services >> >> >>> >> >> >> > 3. A single system of record for all IT processes >> >> >>> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> >> >> > _______________________________________________ >> >> >>> >> >> >> > Kaldi-users mailing list >> >> >>> >> >> >> > Kal...@li... >> >> >>> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> >> >> > >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> ------------------------------------------------------------------------------ >> >> >>> >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >>> >> >> >> 1. A cloud service to automate IT design, transition and >> >> >>> >> >> >> operations >> >> >>> >> >> >> 2. Dashboards that offer high-level views of enterprise >> >> >>> >> >> >> services >> >> >>> >> >> >> 3. A single system of record for all IT processes >> >> >>> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> >> >> _______________________________________________ >> >> >>> >> >> >> Kaldi-users mailing list >> >> >>> >> >> >> Kal...@li... >> >> >>> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > >> >> >>> >> >> > ------------------------------------------------------------------------------ >> >> >>> >> >> > How ServiceNow helps IT people transform IT departments: >> >> >>> >> >> > 1. A cloud service to automate IT design, transition and >> >> >>> >> >> > operations >> >> >>> >> >> > 2. Dashboards that offer high-level views of enterprise >> >> >>> >> >> > services >> >> >>> >> >> > 3. A single system of record for all IT processes >> >> >>> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> >> > _______________________________________________ >> >> >>> >> >> > Kaldi-users mailing list >> >> >>> >> >> > Kal...@li... >> >> >>> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> >> > >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> ------------------------------------------------------------------------------ >> >> >>> >> >> How ServiceNow helps IT people transform IT departments: >> >> >>> >> >> 1. A cloud service to automate IT design, transition and >> >> >>> >> >> operations >> >> >>> >> >> 2. Dashboards that offer high-level views of enterprise >> >> >>> >> >> services >> >> >>> >> >> 3. A single system of record for all IT processes >> >> >>> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> >> _______________________________________________ >> >> >>> >> >> Kaldi-users mailing list >> >> >>> >> >> Kal...@li... >> >> >>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > >> >> >>> >> > ------------------------------------------------------------------------------ >> >> >>> >> > How ServiceNow helps IT people transform IT departments: >> >> >>> >> > 1. A cloud service to automate IT design, transition and >> >> >>> >> > operations >> >> >>> >> > 2. Dashboards that offer high-level views of enterprise >> >> >>> >> > services >> >> >>> >> > 3. A single system of record for all IT processes >> >> >>> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> > _______________________________________________ >> >> >>> >> > Kaldi-users mailing list >> >> >>> >> > Kal...@li... >> >> >>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> > >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> ------------------------------------------------------------------------------ >> >> >>> >> How ServiceNow helps IT people transform IT departments: >> >> >>> >> 1. A cloud service to automate IT design, transition and >> >> >>> >> operations >> >> >>> >> 2. Dashboards that offer high-level views of enterprise services >> >> >>> >> 3. A single system of record for all IT processes >> >> >>> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> >> _______________________________________________ >> >> >>> >> Kaldi-users mailing list >> >> >>> >> Kal...@li... >> >> >>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> >> >> >> >>> > >> >> >>> > >> >> >>> > >> >> >>> > >> >> >>> > >> >> >>> > ------------------------------------------------------------------------------ >> >> >>> > How ServiceNow helps IT people transform IT departments: >> >> >>> > 1. A cloud service to automate IT design, transition and >> >> >>> > operations >> >> >>> > 2. Dashboards that offer high-level views of enterprise services >> >> >>> > 3. A single system of record for all IT processes >> >> >>> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> > _______________________________________________ >> >> >>> > Kaldi-users mailing list >> >> >>> > Kal...@li... >> >> >>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >>> > >> >> >>> >> >> >>> >> >> >>> >> >> >>> >> >> >>> ------------------------------------------------------------------------------ >> >> >>> How ServiceNow helps IT people transform IT departments: >> >> >>> 1. A cloud service to automate IT design, transition and operations >> >> >>> 2. Dashboards that offer high-level views of enterprise services >> >> >>> 3. A single system of record for all IT processes >> >> >>> http://p.sf.net/sfu/servicenow-d2d-j >> >> >>> _______________________________________________ >> >> >>> Kaldi-users mailing list >> >> >>> Kal...@li... >> >> >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >> 1. A cloud service to automate IT design, transition and operations >> >> >> 2. Dashboards that offer high-level views of enterprise services >> >> >> 3. A single system of record for all IT processes >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >> _______________________________________________ >> >> >> Kaldi-users mailing list >> >> >> Kal...@li... >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ >> >> > How ServiceNow helps IT people transform IT departments: >> >> > 1. A cloud service to automate IT design, transition and operations >> >> > 2. Dashboards that offer high-level views of enterprise services >> >> > 3. A single system of record for all IT processes >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> > _______________________________________________ >> >> > Kaldi-users mailing list >> >> > Kal...@li... >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> > >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> How ServiceNow helps IT people transform IT departments: >> >> 1. A cloud service to automate IT design, transition and operations >> >> 2. Dashboards that offer high-level views of enterprise services >> >> 3. A single system of record for all IT processes >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> _______________________________________________ >> >> Kaldi-users mailing list >> >> Kal...@li... >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > >> > >> > >> > >> > ------------------------------------------------------------------------------ >> > How ServiceNow helps IT people transform IT departments: >> > 1. A cloud service to automate IT design, transition and operations >> > 2. Dashboards that offer high-level views of enterprise services >> > 3. A single system of record for all IT processes >> > http://p.sf.net/sfu/servicenow-d2d-j >> > _______________________________________________ >> > Kaldi-users mailing list >> > Kal...@li... >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > >> >> >> ------------------------------------------------------------------------------ >> How ServiceNow helps IT people transform IT departments: >> 1. A cloud service to automate IT design, transition and operations >> 2. Dashboards that offer high-level views of enterprise services >> 3. A single system of record for all IT processes >> http://p.sf.net/sfu/servicenow-d2d-j >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 17:15:57
|
Thanks [arnab, I assume]. I just found the problem myself also. Sorry. Dan On Tue, Jun 4, 2013 at 1:08 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Found the problem! It's in local/score.sh -- in rev 2443 you added an > option for using word insertion penalty before which the lattice > acoustic scores get scaled by lm weight. But the lattice-best-path is > also using a scaling, and so the scores are being scaled twice. The > fix is in the repository. But I haven't checked if the word insertion > penalty has been added to other scoring scripts as well. > > Hao, please rerun and you should get something around 34.76% for dev93. > > On Tue, Jun 4, 2013 at 5:48 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > BTW, I suspect local/make_lexicon_fst.pl > > I made a change which I thought was a simple bug fix that would not > affect > > anything. > > Dan > > > > > > > > On Tue, Jun 4, 2013 at 12:33 PM, Mailing list used for User Communication > > and Updates <kal...@li...> wrote: > >> > >> There seems to be something broken with the current WSJ setup. I reran > >> the setup today and I am getting similar results as Hao. For example, > >> an older setup gets 34.98% on dev93, while the system trained today > >> gets 47.79%. The problem is not features, since the older model with > >> the new features is 34.76%. The problem is also not the dict/lm since > >> the old model with new dict/lm and old features gets 34.98%. It is > >> unlikely to be the new model, since the log-likelihood numbers are in > >> the same ballpark. My current best guess is that some setting in the > >> decode/scoring scripts has changed, which is causing the problem. I am > >> looking into this in more detail. > >> -Arnab > >> > >> On Tue, Jun 4, 2013 at 8:02 AM, Mailing list used for User > >> Communication and Updates <kal...@li...> wrote: > >> > Here are the log for the last update with different number of jobs. I > >> > don't see much of a difference. > >> > > >> > In terms of likelihood shown below, they are pretty consistent. > >> > > >> > nj=2 -109.265 -> -95.5657 > >> > nj=4 -109.265 -> -95.5599 > >> > nj=8 -109.265 -> -95.556 > >> > > >> > Hao > >> > > >> > On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User > >> > Communication and Updates <kal...@li...> wrote: > >> >> OK, I have an idea what might have happened. It's possible that > >> >> because you > >> >> are using such a small number of parallel jobs (2), there is a > problem > >> >> due > >> >> to numerical roundoff. This should not be the case because we > >> >> accumulate in > >> >> double, but to see whether this is a factor, could you please run the > >> >> training with more jobs? E.g. 4 or 8? See if the likelihoods change. > >> >> Also, > >> >> show me the full log from the last update, both before and after you > >> >> make > >> >> the change. > >> >> Dan > >> >> > >> >> > >> >> > >> >> On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User > >> >> Communication and > >> >> Updates <kal...@li...> wrote: > >> >>> > >> >>> They are almost identical. > >> >>> > >> >>> 011c0201 [ > >> >>> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 > >> >>> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 > >> >>> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 > -12.17628 > >> >>> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 > >> >>> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 > -3.351704 > >> >>> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 > >> >>> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 > >> >>> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > >> >>> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 > -9.91317 > >> >>> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 > >> >>> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 > -8.361964 > >> >>> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > >> >>> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 > -3.151051 > >> >>> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 > >> >>> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 > >> >>> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 > >> >>> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 > -10.80855 > >> >>> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 > >> >>> > >> >>> Hao > >> >>> > >> >>> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User > >> >>> Communication and Updates <kal...@li...> > wrote: > >> >>> > OK, everything there looks pretty normal. Let's check the > features. > >> >>> > Try > >> >>> > doing > >> >>> > > >> >>> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head > >> >>> > copy-feats scp:data/train_si284/feats.scp ark,t:- > >> >>> > 011c0201 [ > >> >>> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 > -5.20842 > >> >>> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 > >> >>> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 > >> >>> > -12.17628 > >> >>> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 > >> >>> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 > >> >>> > -3.351702 > >> >>> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 > >> >>> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 > -10.907 > >> >>> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > >> >>> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 > -9.913167 > >> >>> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 > >> >>> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 > >> >>> > -8.361964 > >> >>> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > >> >>> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 > >> >>> > -3.151051 > >> >>> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 > >> >>> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 > -1.814186 > >> >>> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 > >> >>> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 > >> >>> > -10.80855 > >> >>> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 > >> >>> > > >> >>> > and show me the otuput > >> >>> > Dan > >> >>> > > >> >>> > > >> >>> > > >> >>> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User > >> >>> > Communication > >> >>> > and > >> >>> > Updates <kal...@li...> wrote: > >> >>> >> > >> >>> >> Here is the result for > >> >>> >> > >> >>> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >> >>> >> exp/mono0a/decode/log > >> >>> >> > >> >>> >> Hao > >> >>> >> > >> >>> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User > >> >>> >> Communication and Updates <kal...@li...> > >> >>> >> wrote: > >> >>> >> > My corresponding numbers are -108.535 -> -95.116, and the > >> >>> >> > difference > >> >>> >> > is > >> >>> >> > about the same. [It's strange that the absolute numbers differ > >> >>> >> > slightly, > >> >>> >> > though.] > >> >>> >> > Could you do > >> >>> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >> >>> >> > exp/mono0a/decode/log > >> >>> >> > and send the output? > >> >>> >> > > >> >>> >> > Dan > >> >>> >> > > >> >>> >> > > >> >>> >> > > >> >>> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User > >> >>> >> > Communication > >> >>> >> > and > >> >>> >> > Updates <kal...@li...> wrote: > >> >>> >> >> > >> >>> >> >> Copied from exp/mono0a/log/update.39.log > >> >>> >> >> > >> >>> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 > stats, > >> >>> >> >> total > >> >>> >> >> count 849568, avg like/frame -95.5657 > >> >>> >> >> > >> >>> >> >> Hao > >> >>> >> >> > >> >>> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User > >> >>> >> >> Communication and Updates <kal...@li...> > >> >>> >> >> wrote: > >> >>> >> >> > OK, that is quite weird. What is the avg like/frame on the > >> >>> >> >> > final > >> >>> >> >> > update.??.log? > >> >>> >> >> > Dan > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User > >> >>> >> >> > Communication > >> >>> >> >> > and > >> >>> >> >> > Updates <kal...@li...> wrote: > >> >>> >> >> >> > >> >>> >> >> >> Unfortunately those are absolute differences. Here is what > I > >> >>> >> >> >> get > >> >>> >> >> >> for > >> >>> >> >> >> the > >> >>> >> >> >> mono0a. > >> >>> >> >> >> > >> >>> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] > >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 > >> >>> >> >> >> > >> >>> >> >> >> And below is the reported one in RESULTS. > >> >>> >> >> >> > >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 > / > >> >>> >> >> >> 5643, > >> >>> >> >> >> 149 > >> >>> >> >> >> ins, 191 del, 1101 sub ] > >> >>> >> >> >> > >> >>> >> >> >> I will try to checkout the same version as the RESULTS file > >> >>> >> >> >> and > >> >>> >> >> >> run > >> >>> >> >> >> the experiments again. > >> >>> >> >> >> > >> >>> >> >> >> Here is the total data count from > >> >>> >> >> >> exp/mono0a/log/update.0.log. > >> >>> >> >> >> > >> >>> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 > >> >>> >> >> >> stats, > >> >>> >> >> >> total > >> >>> >> >> >> count 849568, avg like/frame -109.265 > >> >>> >> >> >> > >> >>> >> >> >> Thanks for your help. > >> >>> >> >> >> > >> >>> >> >> >> Hao > >> >>> >> >> >> > >> >>> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User > >> >>> >> >> >> Communication and Updates < > kal...@li...> > >> >>> >> >> >> wrote: > >> >>> >> >> >> > Assuming those are relative numbers, they're probably > >> >>> >> >> >> > within > >> >>> >> >> >> > the > >> >>> >> >> >> > margin > >> >>> >> >> >> > of > >> >>> >> >> >> > error. See if it's the same on other test sets. Because > >> >>> >> >> >> > various > >> >>> >> >> >> > algorithms > >> >>> >> >> >> > call rand(), and different machines implement this > >> >>> >> >> >> > differently, > >> >>> >> >> >> > results > >> >>> >> >> >> > aren't fully reproducible. > >> >>> >> >> >> > > >> >>> >> >> >> > But to check that it's not some code change that hurts > the > >> >>> >> >> >> > results, > >> >>> >> >> >> > you > >> >>> >> >> >> > could try checking out a copy with the same revision > number > >> >>> >> >> >> > as > >> >>> >> >> >> > the > >> >>> >> >> >> > RESULTS > >> >>> >> >> >> > file, and running again and seeing what results you get. > >> >>> >> >> >> > If > >> >>> >> >> >> > there > >> >>> >> >> >> > is > >> >>> >> >> >> > a > >> >>> >> >> >> > difference, I'd like to know. > >> >>> >> >> >> > > >> >>> >> >> >> > It's also possible some files were missing in your WSJ > >> >>> >> >> >> > distribution-- > >> >>> >> >> >> > let me > >> >>> >> >> >> > know the data count reported in one the */log/update.log > >> >>> >> >> >> > files > >> >>> >> >> >> > and > >> >>> >> >> >> > I'll > >> >>> >> >> >> > compare with a local copy. > >> >>> >> >> >> > > >> >>> >> >> >> > Dan > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for > User > >> >>> >> >> >> > Communication > >> >>> >> >> >> > and > >> >>> >> >> >> > Updates <kal...@li...> wrote: > >> >>> >> >> >> >> > >> >>> >> >> >> >> Hi, > >> >>> >> >> >> >> > >> >>> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to > >> >>> >> >> >> >> reproduce > >> >>> >> >> >> >> the > >> >>> >> >> >> >> results stated in the RESULTS file. Are we supposed to > get > >> >>> >> >> >> >> exactly > >> >>> >> >> >> >> the > >> >>> >> >> >> >> same numbers with the (hyper-)parameters given in the > >> >>> >> >> >> >> recipe? > >> >>> >> >> >> >> > >> >>> >> >> >> >> Just to give a sense of how far I'm off. My monophone > >> >>> >> >> >> >> models > >> >>> >> >> >> >> perform > >> >>> >> >> >> >> 10% > >> >>> >> >> >> >> worse than the reported results on eval92 and the > >> >>> >> >> >> >> resulting > >> >>> >> >> >> >> triphone > >> >>> >> >> >> >> models perform 4% worse. > >> >>> >> >> >> >> > >> >>> >> >> >> >> Thanks, > >> >>> >> >> >> >> Hao > >> >>> >> >> >> >> > >> >>> >> >> >> >> > >> >>> >> >> >> >> > >> >>> >> >> >> >> > >> >>> >> >> >> >> > >> >>> >> >> >> >> > >> >>> >> >> >> >> > >> >>> >> >> >> >> > ------------------------------------------------------------------------------ > >> >>> >> >> >> >> How ServiceNow helps IT people transform IT departments: > >> >>> >> >> >> >> 1. A cloud service to automate IT design, transition and > >> >>> >> >> >> >> operations > >> >>> >> >> >> >> 2. Dashboards that offer high-level views of enterprise > >> >>> >> >> >> >> services > >> >>> >> >> >> >> 3. A single system of record for all IT processes > >> >>> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> >> >> >> _______________________________________________ > >> >>> >> >> >> >> Kaldi-users mailing list > >> >>> >> >> >> >> Kal...@li... > >> >>> >> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > >> >>> >> >> >> > > ------------------------------------------------------------------------------ > >> >>> >> >> >> > How ServiceNow helps IT people transform IT departments: > >> >>> >> >> >> > 1. A cloud service to automate IT design, transition and > >> >>> >> >> >> > operations > >> >>> >> >> >> > 2. Dashboards that offer high-level views of enterprise > >> >>> >> >> >> > services > >> >>> >> >> >> > 3. A single system of record for all IT processes > >> >>> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> >> >> > _______________________________________________ > >> >>> >> >> >> > Kaldi-users mailing list > >> >>> >> >> >> > Kal...@li... > >> >>> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> >> >> > > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > ------------------------------------------------------------------------------ > >> >>> >> >> >> How ServiceNow helps IT people transform IT departments: > >> >>> >> >> >> 1. A cloud service to automate IT design, transition and > >> >>> >> >> >> operations > >> >>> >> >> >> 2. Dashboards that offer high-level views of enterprise > >> >>> >> >> >> services > >> >>> >> >> >> 3. A single system of record for all IT processes > >> >>> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> >> >> _______________________________________________ > >> >>> >> >> >> Kaldi-users mailing list > >> >>> >> >> >> Kal...@li... > >> >>> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > >> >>> >> >> > > ------------------------------------------------------------------------------ > >> >>> >> >> > How ServiceNow helps IT people transform IT departments: > >> >>> >> >> > 1. A cloud service to automate IT design, transition and > >> >>> >> >> > operations > >> >>> >> >> > 2. Dashboards that offer high-level views of enterprise > >> >>> >> >> > services > >> >>> >> >> > 3. A single system of record for all IT processes > >> >>> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> >> > _______________________________________________ > >> >>> >> >> > Kaldi-users mailing list > >> >>> >> >> > Kal...@li... > >> >>> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> >> > > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > ------------------------------------------------------------------------------ > >> >>> >> >> How ServiceNow helps IT people transform IT departments: > >> >>> >> >> 1. A cloud service to automate IT design, transition and > >> >>> >> >> operations > >> >>> >> >> 2. Dashboards that offer high-level views of enterprise > services > >> >>> >> >> 3. A single system of record for all IT processes > >> >>> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> >> _______________________________________________ > >> >>> >> >> Kaldi-users mailing list > >> >>> >> >> Kal...@li... > >> >>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> > > >> >>> >> > > >> >>> >> > > >> >>> >> > > >> >>> >> > > >> >>> >> > > >> >>> >> > > ------------------------------------------------------------------------------ > >> >>> >> > How ServiceNow helps IT people transform IT departments: > >> >>> >> > 1. A cloud service to automate IT design, transition and > >> >>> >> > operations > >> >>> >> > 2. Dashboards that offer high-level views of enterprise > services > >> >>> >> > 3. A single system of record for all IT processes > >> >>> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> > _______________________________________________ > >> >>> >> > Kaldi-users mailing list > >> >>> >> > Kal...@li... > >> >>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> > > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > ------------------------------------------------------------------------------ > >> >>> >> How ServiceNow helps IT people transform IT departments: > >> >>> >> 1. A cloud service to automate IT design, transition and > operations > >> >>> >> 2. Dashboards that offer high-level views of enterprise services > >> >>> >> 3. A single system of record for all IT processes > >> >>> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >>> >> _______________________________________________ > >> >>> >> Kaldi-users mailing list > >> >>> >> Kal...@li... > >> >>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> >> > >> >>> > > >> >>> > > >> >>> > > >> >>> > > >> >>> > > ------------------------------------------------------------------------------ > >> >>> > How ServiceNow helps IT people transform IT departments: > >> >>> > 1. A cloud service to automate IT design, transition and > operations > >> >>> > 2. Dashboards that offer high-level views of enterprise services > >> >>> > 3. A single system of record for all IT processes > >> >>> > http://p.sf.net/sfu/servicenow-d2d-j > >> >>> > _______________________________________________ > >> >>> > Kaldi-users mailing list > >> >>> > Kal...@li... > >> >>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >>> > > >> >>> > >> >>> > >> >>> > >> >>> > ------------------------------------------------------------------------------ > >> >>> How ServiceNow helps IT people transform IT departments: > >> >>> 1. A cloud service to automate IT design, transition and operations > >> >>> 2. Dashboards that offer high-level views of enterprise services > >> >>> 3. A single system of record for all IT processes > >> >>> http://p.sf.net/sfu/servicenow-d2d-j > >> >>> _______________________________________________ > >> >>> Kaldi-users mailing list > >> >>> Kal...@li... > >> >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ > >> >> How ServiceNow helps IT people transform IT departments: > >> >> 1. A cloud service to automate IT design, transition and operations > >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> 3. A single system of record for all IT processes > >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> _______________________________________________ > >> >> Kaldi-users mailing list > >> >> Kal...@li... > >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > >> > > >> > > >> > > ------------------------------------------------------------------------------ > >> > How ServiceNow helps IT people transform IT departments: > >> > 1. A cloud service to automate IT design, transition and operations > >> > 2. Dashboards that offer high-level views of enterprise services > >> > 3. A single system of record for all IT processes > >> > http://p.sf.net/sfu/servicenow-d2d-j > >> > _______________________________________________ > >> > Kaldi-users mailing list > >> > Kal...@li... > >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 17:08:54
|
Found the problem! It's in local/score.sh -- in rev 2443 you added an option for using word insertion penalty before which the lattice acoustic scores get scaled by lm weight. But the lattice-best-path is also using a scaling, and so the scores are being scaled twice. The fix is in the repository. But I haven't checked if the word insertion penalty has been added to other scoring scripts as well. Hao, please rerun and you should get something around 34.76% for dev93. On Tue, Jun 4, 2013 at 5:48 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > BTW, I suspect local/make_lexicon_fst.pl > I made a change which I thought was a simple bug fix that would not affect > anything. > Dan > > > > On Tue, Jun 4, 2013 at 12:33 PM, Mailing list used for User Communication > and Updates <kal...@li...> wrote: >> >> There seems to be something broken with the current WSJ setup. I reran >> the setup today and I am getting similar results as Hao. For example, >> an older setup gets 34.98% on dev93, while the system trained today >> gets 47.79%. The problem is not features, since the older model with >> the new features is 34.76%. The problem is also not the dict/lm since >> the old model with new dict/lm and old features gets 34.98%. It is >> unlikely to be the new model, since the log-likelihood numbers are in >> the same ballpark. My current best guess is that some setting in the >> decode/scoring scripts has changed, which is causing the problem. I am >> looking into this in more detail. >> -Arnab >> >> On Tue, Jun 4, 2013 at 8:02 AM, Mailing list used for User >> Communication and Updates <kal...@li...> wrote: >> > Here are the log for the last update with different number of jobs. I >> > don't see much of a difference. >> > >> > In terms of likelihood shown below, they are pretty consistent. >> > >> > nj=2 -109.265 -> -95.5657 >> > nj=4 -109.265 -> -95.5599 >> > nj=8 -109.265 -> -95.556 >> > >> > Hao >> > >> > On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User >> > Communication and Updates <kal...@li...> wrote: >> >> OK, I have an idea what might have happened. It's possible that >> >> because you >> >> are using such a small number of parallel jobs (2), there is a problem >> >> due >> >> to numerical roundoff. This should not be the case because we >> >> accumulate in >> >> double, but to see whether this is a factor, could you please run the >> >> training with more jobs? E.g. 4 or 8? See if the likelihoods change. >> >> Also, >> >> show me the full log from the last update, both before and after you >> >> make >> >> the change. >> >> Dan >> >> >> >> >> >> >> >> On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User >> >> Communication and >> >> Updates <kal...@li...> wrote: >> >>> >> >>> They are almost identical. >> >>> >> >>> 011c0201 [ >> >>> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 >> >>> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 >> >>> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 -12.17628 >> >>> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 >> >>> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 -3.351704 >> >>> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 >> >>> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >> >>> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >> >>> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 -9.91317 >> >>> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 >> >>> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 >> >>> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >> >>> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 -3.151051 >> >>> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 >> >>> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 >> >>> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 >> >>> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 -10.80855 >> >>> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 >> >>> >> >>> Hao >> >>> >> >>> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User >> >>> Communication and Updates <kal...@li...> wrote: >> >>> > OK, everything there looks pretty normal. Let's check the features. >> >>> > Try >> >>> > doing >> >>> > >> >>> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head >> >>> > copy-feats scp:data/train_si284/feats.scp ark,t:- >> >>> > 011c0201 [ >> >>> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 >> >>> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 >> >>> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 >> >>> > -12.17628 >> >>> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 >> >>> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 >> >>> > -3.351702 >> >>> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 >> >>> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >> >>> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >> >>> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 >> >>> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 >> >>> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 >> >>> > -8.361964 >> >>> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >> >>> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 >> >>> > -3.151051 >> >>> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 >> >>> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 >> >>> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 >> >>> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 >> >>> > -10.80855 >> >>> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 >> >>> > >> >>> > and show me the otuput >> >>> > Dan >> >>> > >> >>> > >> >>> > >> >>> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User >> >>> > Communication >> >>> > and >> >>> > Updates <kal...@li...> wrote: >> >>> >> >> >>> >> Here is the result for >> >>> >> >> >>> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> >>> >> exp/mono0a/decode/log >> >>> >> >> >>> >> Hao >> >>> >> >> >>> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User >> >>> >> Communication and Updates <kal...@li...> >> >>> >> wrote: >> >>> >> > My corresponding numbers are -108.535 -> -95.116, and the >> >>> >> > difference >> >>> >> > is >> >>> >> > about the same. [It's strange that the absolute numbers differ >> >>> >> > slightly, >> >>> >> > though.] >> >>> >> > Could you do >> >>> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> >>> >> > exp/mono0a/decode/log >> >>> >> > and send the output? >> >>> >> > >> >>> >> > Dan >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User >> >>> >> > Communication >> >>> >> > and >> >>> >> > Updates <kal...@li...> wrote: >> >>> >> >> >> >>> >> >> Copied from exp/mono0a/log/update.39.log >> >>> >> >> >> >>> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, >> >>> >> >> total >> >>> >> >> count 849568, avg like/frame -95.5657 >> >>> >> >> >> >>> >> >> Hao >> >>> >> >> >> >>> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User >> >>> >> >> Communication and Updates <kal...@li...> >> >>> >> >> wrote: >> >>> >> >> > OK, that is quite weird. What is the avg like/frame on the >> >>> >> >> > final >> >>> >> >> > update.??.log? >> >>> >> >> > Dan >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User >> >>> >> >> > Communication >> >>> >> >> > and >> >>> >> >> > Updates <kal...@li...> wrote: >> >>> >> >> >> >> >>> >> >> >> Unfortunately those are absolute differences. Here is what I >> >>> >> >> >> get >> >>> >> >> >> for >> >>> >> >> >> the >> >>> >> >> >> mono0a. >> >>> >> >> >> >> >>> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 >> >>> >> >> >> >> >>> >> >> >> And below is the reported one in RESULTS. >> >>> >> >> >> >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / >> >>> >> >> >> 5643, >> >>> >> >> >> 149 >> >>> >> >> >> ins, 191 del, 1101 sub ] >> >>> >> >> >> >> >>> >> >> >> I will try to checkout the same version as the RESULTS file >> >>> >> >> >> and >> >>> >> >> >> run >> >>> >> >> >> the experiments again. >> >>> >> >> >> >> >>> >> >> >> Here is the total data count from >> >>> >> >> >> exp/mono0a/log/update.0.log. >> >>> >> >> >> >> >>> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 >> >>> >> >> >> stats, >> >>> >> >> >> total >> >>> >> >> >> count 849568, avg like/frame -109.265 >> >>> >> >> >> >> >>> >> >> >> Thanks for your help. >> >>> >> >> >> >> >>> >> >> >> Hao >> >>> >> >> >> >> >>> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User >> >>> >> >> >> Communication and Updates <kal...@li...> >> >>> >> >> >> wrote: >> >>> >> >> >> > Assuming those are relative numbers, they're probably >> >>> >> >> >> > within >> >>> >> >> >> > the >> >>> >> >> >> > margin >> >>> >> >> >> > of >> >>> >> >> >> > error. See if it's the same on other test sets. Because >> >>> >> >> >> > various >> >>> >> >> >> > algorithms >> >>> >> >> >> > call rand(), and different machines implement this >> >>> >> >> >> > differently, >> >>> >> >> >> > results >> >>> >> >> >> > aren't fully reproducible. >> >>> >> >> >> > >> >>> >> >> >> > But to check that it's not some code change that hurts the >> >>> >> >> >> > results, >> >>> >> >> >> > you >> >>> >> >> >> > could try checking out a copy with the same revision number >> >>> >> >> >> > as >> >>> >> >> >> > the >> >>> >> >> >> > RESULTS >> >>> >> >> >> > file, and running again and seeing what results you get. >> >>> >> >> >> > If >> >>> >> >> >> > there >> >>> >> >> >> > is >> >>> >> >> >> > a >> >>> >> >> >> > difference, I'd like to know. >> >>> >> >> >> > >> >>> >> >> >> > It's also possible some files were missing in your WSJ >> >>> >> >> >> > distribution-- >> >>> >> >> >> > let me >> >>> >> >> >> > know the data count reported in one the */log/update.log >> >>> >> >> >> > files >> >>> >> >> >> > and >> >>> >> >> >> > I'll >> >>> >> >> >> > compare with a local copy. >> >>> >> >> >> > >> >>> >> >> >> > Dan >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User >> >>> >> >> >> > Communication >> >>> >> >> >> > and >> >>> >> >> >> > Updates <kal...@li...> wrote: >> >>> >> >> >> >> >> >>> >> >> >> >> Hi, >> >>> >> >> >> >> >> >>> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to >> >>> >> >> >> >> reproduce >> >>> >> >> >> >> the >> >>> >> >> >> >> results stated in the RESULTS file. Are we supposed to get >> >>> >> >> >> >> exactly >> >>> >> >> >> >> the >> >>> >> >> >> >> same numbers with the (hyper-)parameters given in the >> >>> >> >> >> >> recipe? >> >>> >> >> >> >> >> >>> >> >> >> >> Just to give a sense of how far I'm off. My monophone >> >>> >> >> >> >> models >> >>> >> >> >> >> perform >> >>> >> >> >> >> 10% >> >>> >> >> >> >> worse than the reported results on eval92 and the >> >>> >> >> >> >> resulting >> >>> >> >> >> >> triphone >> >>> >> >> >> >> models perform 4% worse. >> >>> >> >> >> >> >> >>> >> >> >> >> Thanks, >> >>> >> >> >> >> Hao >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> >> >>> >> >> >> >> ------------------------------------------------------------------------------ >> >>> >> >> >> >> How ServiceNow helps IT people transform IT departments: >> >>> >> >> >> >> 1. A cloud service to automate IT design, transition and >> >>> >> >> >> >> operations >> >>> >> >> >> >> 2. Dashboards that offer high-level views of enterprise >> >>> >> >> >> >> services >> >>> >> >> >> >> 3. A single system of record for all IT processes >> >>> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> >> >> >> _______________________________________________ >> >>> >> >> >> >> Kaldi-users mailing list >> >>> >> >> >> >> Kal...@li... >> >>> >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > >> >>> >> >> >> > ------------------------------------------------------------------------------ >> >>> >> >> >> > How ServiceNow helps IT people transform IT departments: >> >>> >> >> >> > 1. A cloud service to automate IT design, transition and >> >>> >> >> >> > operations >> >>> >> >> >> > 2. Dashboards that offer high-level views of enterprise >> >>> >> >> >> > services >> >>> >> >> >> > 3. A single system of record for all IT processes >> >>> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> >> >> > _______________________________________________ >> >>> >> >> >> > Kaldi-users mailing list >> >>> >> >> >> > Kal...@li... >> >>> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> >> >> > >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> ------------------------------------------------------------------------------ >> >>> >> >> >> How ServiceNow helps IT people transform IT departments: >> >>> >> >> >> 1. A cloud service to automate IT design, transition and >> >>> >> >> >> operations >> >>> >> >> >> 2. Dashboards that offer high-level views of enterprise >> >>> >> >> >> services >> >>> >> >> >> 3. A single system of record for all IT processes >> >>> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> >> >> _______________________________________________ >> >>> >> >> >> Kaldi-users mailing list >> >>> >> >> >> Kal...@li... >> >>> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > >> >>> >> >> > ------------------------------------------------------------------------------ >> >>> >> >> > How ServiceNow helps IT people transform IT departments: >> >>> >> >> > 1. A cloud service to automate IT design, transition and >> >>> >> >> > operations >> >>> >> >> > 2. Dashboards that offer high-level views of enterprise >> >>> >> >> > services >> >>> >> >> > 3. A single system of record for all IT processes >> >>> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> >> > _______________________________________________ >> >>> >> >> > Kaldi-users mailing list >> >>> >> >> > Kal...@li... >> >>> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> >> > >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> ------------------------------------------------------------------------------ >> >>> >> >> How ServiceNow helps IT people transform IT departments: >> >>> >> >> 1. A cloud service to automate IT design, transition and >> >>> >> >> operations >> >>> >> >> 2. Dashboards that offer high-level views of enterprise services >> >>> >> >> 3. A single system of record for all IT processes >> >>> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> >> _______________________________________________ >> >>> >> >> Kaldi-users mailing list >> >>> >> >> Kal...@li... >> >>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > >> >>> >> > ------------------------------------------------------------------------------ >> >>> >> > How ServiceNow helps IT people transform IT departments: >> >>> >> > 1. A cloud service to automate IT design, transition and >> >>> >> > operations >> >>> >> > 2. Dashboards that offer high-level views of enterprise services >> >>> >> > 3. A single system of record for all IT processes >> >>> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> > _______________________________________________ >> >>> >> > Kaldi-users mailing list >> >>> >> > Kal...@li... >> >>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> > >> >>> >> >> >>> >> >> >>> >> >> >>> >> >> >>> >> ------------------------------------------------------------------------------ >> >>> >> How ServiceNow helps IT people transform IT departments: >> >>> >> 1. A cloud service to automate IT design, transition and operations >> >>> >> 2. Dashboards that offer high-level views of enterprise services >> >>> >> 3. A single system of record for all IT processes >> >>> >> http://p.sf.net/sfu/servicenow-d2d-j >> >>> >> _______________________________________________ >> >>> >> Kaldi-users mailing list >> >>> >> Kal...@li... >> >>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> >> >> >>> > >> >>> > >> >>> > >> >>> > >> >>> > ------------------------------------------------------------------------------ >> >>> > How ServiceNow helps IT people transform IT departments: >> >>> > 1. A cloud service to automate IT design, transition and operations >> >>> > 2. Dashboards that offer high-level views of enterprise services >> >>> > 3. A single system of record for all IT processes >> >>> > http://p.sf.net/sfu/servicenow-d2d-j >> >>> > _______________________________________________ >> >>> > Kaldi-users mailing list >> >>> > Kal...@li... >> >>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >>> > >> >>> >> >>> >> >>> >> >>> ------------------------------------------------------------------------------ >> >>> How ServiceNow helps IT people transform IT departments: >> >>> 1. A cloud service to automate IT design, transition and operations >> >>> 2. Dashboards that offer high-level views of enterprise services >> >>> 3. A single system of record for all IT processes >> >>> http://p.sf.net/sfu/servicenow-d2d-j >> >>> _______________________________________________ >> >>> Kaldi-users mailing list >> >>> Kal...@li... >> >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> How ServiceNow helps IT people transform IT departments: >> >> 1. A cloud service to automate IT design, transition and operations >> >> 2. Dashboards that offer high-level views of enterprise services >> >> 3. A single system of record for all IT processes >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> _______________________________________________ >> >> Kaldi-users mailing list >> >> Kal...@li... >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> > >> > >> > ------------------------------------------------------------------------------ >> > How ServiceNow helps IT people transform IT departments: >> > 1. A cloud service to automate IT design, transition and operations >> > 2. Dashboards that offer high-level views of enterprise services >> > 3. A single system of record for all IT processes >> > http://p.sf.net/sfu/servicenow-d2d-j >> > _______________________________________________ >> > Kaldi-users mailing list >> > Kal...@li... >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > >> >> >> ------------------------------------------------------------------------------ >> How ServiceNow helps IT people transform IT departments: >> 1. A cloud service to automate IT design, transition and operations >> 2. Dashboards that offer high-level views of enterprise services >> 3. A single system of record for all IT processes >> http://p.sf.net/sfu/servicenow-d2d-j >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 16:48:43
|
BTW, I suspect local/make_lexicon_fst.pl I made a change which I thought was a simple bug fix that would not affect anything. Dan On Tue, Jun 4, 2013 at 12:33 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > There seems to be something broken with the current WSJ setup. I reran > the setup today and I am getting similar results as Hao. For example, > an older setup gets 34.98% on dev93, while the system trained today > gets 47.79%. The problem is not features, since the older model with > the new features is 34.76%. The problem is also not the dict/lm since > the old model with new dict/lm and old features gets 34.98%. It is > unlikely to be the new model, since the log-likelihood numbers are in > the same ballpark. My current best guess is that some setting in the > decode/scoring scripts has changed, which is causing the problem. I am > looking into this in more detail. > -Arnab > > On Tue, Jun 4, 2013 at 8:02 AM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > Here are the log for the last update with different number of jobs. I > > don't see much of a difference. > > > > In terms of likelihood shown below, they are pretty consistent. > > > > nj=2 -109.265 -> -95.5657 > > nj=4 -109.265 -> -95.5599 > > nj=8 -109.265 -> -95.556 > > > > Hao > > > > On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User > > Communication and Updates <kal...@li...> wrote: > >> OK, I have an idea what might have happened. It's possible that > because you > >> are using such a small number of parallel jobs (2), there is a problem > due > >> to numerical roundoff. This should not be the case because we > accumulate in > >> double, but to see whether this is a factor, could you please run the > >> training with more jobs? E.g. 4 or 8? See if the likelihoods change. > Also, > >> show me the full log from the last update, both before and after you > make > >> the change. > >> Dan > >> > >> > >> > >> On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User > Communication and > >> Updates <kal...@li...> wrote: > >>> > >>> They are almost identical. > >>> > >>> 011c0201 [ > >>> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 > >>> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 > >>> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 -12.17628 > >>> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 > >>> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 -3.351704 > >>> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 > >>> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 > >>> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > >>> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 -9.91317 > >>> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 > >>> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 > >>> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > >>> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 -3.151051 > >>> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 > >>> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 > >>> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 > >>> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 -10.80855 > >>> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 > >>> > >>> Hao > >>> > >>> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User > >>> Communication and Updates <kal...@li...> wrote: > >>> > OK, everything there looks pretty normal. Let's check the features. > >>> > Try > >>> > doing > >>> > > >>> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head > >>> > copy-feats scp:data/train_si284/feats.scp ark,t:- > >>> > 011c0201 [ > >>> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 > >>> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 > >>> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 > -12.17628 > >>> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 > >>> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 > -3.351702 > >>> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 > >>> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 > >>> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > >>> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 > >>> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 > >>> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 > >>> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > >>> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 -3.151051 > >>> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 > >>> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 > >>> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 > >>> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 > -10.80855 > >>> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 > >>> > > >>> > and show me the otuput > >>> > Dan > >>> > > >>> > > >>> > > >>> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User > Communication > >>> > and > >>> > Updates <kal...@li...> wrote: > >>> >> > >>> >> Here is the result for > >>> >> > >>> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >>> >> exp/mono0a/decode/log > >>> >> > >>> >> Hao > >>> >> > >>> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User > >>> >> Communication and Updates <kal...@li...> > wrote: > >>> >> > My corresponding numbers are -108.535 -> -95.116, and the > difference > >>> >> > is > >>> >> > about the same. [It's strange that the absolute numbers differ > >>> >> > slightly, > >>> >> > though.] > >>> >> > Could you do > >>> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >>> >> > exp/mono0a/decode/log > >>> >> > and send the output? > >>> >> > > >>> >> > Dan > >>> >> > > >>> >> > > >>> >> > > >>> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User > >>> >> > Communication > >>> >> > and > >>> >> > Updates <kal...@li...> wrote: > >>> >> >> > >>> >> >> Copied from exp/mono0a/log/update.39.log > >>> >> >> > >>> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, > >>> >> >> total > >>> >> >> count 849568, avg like/frame -95.5657 > >>> >> >> > >>> >> >> Hao > >>> >> >> > >>> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User > >>> >> >> Communication and Updates <kal...@li...> > wrote: > >>> >> >> > OK, that is quite weird. What is the avg like/frame on the > final > >>> >> >> > update.??.log? > >>> >> >> > Dan > >>> >> >> > > >>> >> >> > > >>> >> >> > > >>> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User > >>> >> >> > Communication > >>> >> >> > and > >>> >> >> > Updates <kal...@li...> wrote: > >>> >> >> >> > >>> >> >> >> Unfortunately those are absolute differences. Here is what I > get > >>> >> >> >> for > >>> >> >> >> the > >>> >> >> >> mono0a. > >>> >> >> >> > >>> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] > >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 > >>> >> >> >> > >>> >> >> >> And below is the reported one in RESULTS. > >>> >> >> >> > >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / > >>> >> >> >> 5643, > >>> >> >> >> 149 > >>> >> >> >> ins, 191 del, 1101 sub ] > >>> >> >> >> > >>> >> >> >> I will try to checkout the same version as the RESULTS file > and > >>> >> >> >> run > >>> >> >> >> the experiments again. > >>> >> >> >> > >>> >> >> >> Here is the total data count from exp/mono0a/log/update.0.log. > >>> >> >> >> > >>> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 > stats, > >>> >> >> >> total > >>> >> >> >> count 849568, avg like/frame -109.265 > >>> >> >> >> > >>> >> >> >> Thanks for your help. > >>> >> >> >> > >>> >> >> >> Hao > >>> >> >> >> > >>> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User > >>> >> >> >> Communication and Updates <kal...@li...> > >>> >> >> >> wrote: > >>> >> >> >> > Assuming those are relative numbers, they're probably within > >>> >> >> >> > the > >>> >> >> >> > margin > >>> >> >> >> > of > >>> >> >> >> > error. See if it's the same on other test sets. Because > >>> >> >> >> > various > >>> >> >> >> > algorithms > >>> >> >> >> > call rand(), and different machines implement this > differently, > >>> >> >> >> > results > >>> >> >> >> > aren't fully reproducible. > >>> >> >> >> > > >>> >> >> >> > But to check that it's not some code change that hurts the > >>> >> >> >> > results, > >>> >> >> >> > you > >>> >> >> >> > could try checking out a copy with the same revision number > as > >>> >> >> >> > the > >>> >> >> >> > RESULTS > >>> >> >> >> > file, and running again and seeing what results you get. If > >>> >> >> >> > there > >>> >> >> >> > is > >>> >> >> >> > a > >>> >> >> >> > difference, I'd like to know. > >>> >> >> >> > > >>> >> >> >> > It's also possible some files were missing in your WSJ > >>> >> >> >> > distribution-- > >>> >> >> >> > let me > >>> >> >> >> > know the data count reported in one the */log/update.log > files > >>> >> >> >> > and > >>> >> >> >> > I'll > >>> >> >> >> > compare with a local copy. > >>> >> >> >> > > >>> >> >> >> > Dan > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User > >>> >> >> >> > Communication > >>> >> >> >> > and > >>> >> >> >> > Updates <kal...@li...> wrote: > >>> >> >> >> >> > >>> >> >> >> >> Hi, > >>> >> >> >> >> > >>> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to > >>> >> >> >> >> reproduce > >>> >> >> >> >> the > >>> >> >> >> >> results stated in the RESULTS file. Are we supposed to get > >>> >> >> >> >> exactly > >>> >> >> >> >> the > >>> >> >> >> >> same numbers with the (hyper-)parameters given in the > recipe? > >>> >> >> >> >> > >>> >> >> >> >> Just to give a sense of how far I'm off. My monophone > models > >>> >> >> >> >> perform > >>> >> >> >> >> 10% > >>> >> >> >> >> worse than the reported results on eval92 and the resulting > >>> >> >> >> >> triphone > >>> >> >> >> >> models perform 4% worse. > >>> >> >> >> >> > >>> >> >> >> >> Thanks, > >>> >> >> >> >> Hao > >>> >> >> >> >> > >>> >> >> >> >> > >>> >> >> >> >> > >>> >> >> >> >> > >>> >> >> >> >> > >>> >> >> >> >> > >>> >> >> >> >> > ------------------------------------------------------------------------------ > >>> >> >> >> >> How ServiceNow helps IT people transform IT departments: > >>> >> >> >> >> 1. A cloud service to automate IT design, transition and > >>> >> >> >> >> operations > >>> >> >> >> >> 2. Dashboards that offer high-level views of enterprise > >>> >> >> >> >> services > >>> >> >> >> >> 3. A single system of record for all IT processes > >>> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >>> >> >> >> >> _______________________________________________ > >>> >> >> >> >> Kaldi-users mailing list > >>> >> >> >> >> Kal...@li... > >>> >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > >>> >> >> >> > > ------------------------------------------------------------------------------ > >>> >> >> >> > How ServiceNow helps IT people transform IT departments: > >>> >> >> >> > 1. A cloud service to automate IT design, transition and > >>> >> >> >> > operations > >>> >> >> >> > 2. Dashboards that offer high-level views of enterprise > >>> >> >> >> > services > >>> >> >> >> > 3. A single system of record for all IT processes > >>> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >>> >> >> >> > _______________________________________________ > >>> >> >> >> > Kaldi-users mailing list > >>> >> >> >> > Kal...@li... > >>> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> >> >> > > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > ------------------------------------------------------------------------------ > >>> >> >> >> How ServiceNow helps IT people transform IT departments: > >>> >> >> >> 1. A cloud service to automate IT design, transition and > >>> >> >> >> operations > >>> >> >> >> 2. Dashboards that offer high-level views of enterprise > services > >>> >> >> >> 3. A single system of record for all IT processes > >>> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >>> >> >> >> _______________________________________________ > >>> >> >> >> Kaldi-users mailing list > >>> >> >> >> Kal...@li... > >>> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> >> > > >>> >> >> > > >>> >> >> > > >>> >> >> > > >>> >> >> > > >>> >> >> > > >>> >> >> > > ------------------------------------------------------------------------------ > >>> >> >> > How ServiceNow helps IT people transform IT departments: > >>> >> >> > 1. A cloud service to automate IT design, transition and > >>> >> >> > operations > >>> >> >> > 2. Dashboards that offer high-level views of enterprise > services > >>> >> >> > 3. A single system of record for all IT processes > >>> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >>> >> >> > _______________________________________________ > >>> >> >> > Kaldi-users mailing list > >>> >> >> > Kal...@li... > >>> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> >> > > >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > ------------------------------------------------------------------------------ > >>> >> >> How ServiceNow helps IT people transform IT departments: > >>> >> >> 1. A cloud service to automate IT design, transition and > operations > >>> >> >> 2. Dashboards that offer high-level views of enterprise services > >>> >> >> 3. A single system of record for all IT processes > >>> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >>> >> >> _______________________________________________ > >>> >> >> Kaldi-users mailing list > >>> >> >> Kal...@li... > >>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> > > >>> >> > > >>> >> > > >>> >> > > >>> >> > > >>> >> > > ------------------------------------------------------------------------------ > >>> >> > How ServiceNow helps IT people transform IT departments: > >>> >> > 1. A cloud service to automate IT design, transition and > operations > >>> >> > 2. Dashboards that offer high-level views of enterprise services > >>> >> > 3. A single system of record for all IT processes > >>> >> > http://p.sf.net/sfu/servicenow-d2d-j > >>> >> > _______________________________________________ > >>> >> > Kaldi-users mailing list > >>> >> > Kal...@li... > >>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> > > >>> >> > >>> >> > >>> >> > >>> >> > ------------------------------------------------------------------------------ > >>> >> How ServiceNow helps IT people transform IT departments: > >>> >> 1. A cloud service to automate IT design, transition and operations > >>> >> 2. Dashboards that offer high-level views of enterprise services > >>> >> 3. A single system of record for all IT processes > >>> >> http://p.sf.net/sfu/servicenow-d2d-j > >>> >> _______________________________________________ > >>> >> Kaldi-users mailing list > >>> >> Kal...@li... > >>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> >> > >>> > > >>> > > >>> > > >>> > > ------------------------------------------------------------------------------ > >>> > How ServiceNow helps IT people transform IT departments: > >>> > 1. A cloud service to automate IT design, transition and operations > >>> > 2. Dashboards that offer high-level views of enterprise services > >>> > 3. A single system of record for all IT processes > >>> > http://p.sf.net/sfu/servicenow-d2d-j > >>> > _______________________________________________ > >>> > Kaldi-users mailing list > >>> > Kal...@li... > >>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >>> > > >>> > >>> > >>> > ------------------------------------------------------------------------------ > >>> How ServiceNow helps IT people transform IT departments: > >>> 1. A cloud service to automate IT design, transition and operations > >>> 2. Dashboards that offer high-level views of enterprise services > >>> 3. A single system of record for all IT processes > >>> http://p.sf.net/sfu/servicenow-d2d-j > >>> _______________________________________________ > >>> Kaldi-users mailing list > >>> Kal...@li... > >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 16:33:43
|
There seems to be something broken with the current WSJ setup. I reran the setup today and I am getting similar results as Hao. For example, an older setup gets 34.98% on dev93, while the system trained today gets 47.79%. The problem is not features, since the older model with the new features is 34.76%. The problem is also not the dict/lm since the old model with new dict/lm and old features gets 34.98%. It is unlikely to be the new model, since the log-likelihood numbers are in the same ballpark. My current best guess is that some setting in the decode/scoring scripts has changed, which is causing the problem. I am looking into this in more detail. -Arnab On Tue, Jun 4, 2013 at 8:02 AM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Here are the log for the last update with different number of jobs. I > don't see much of a difference. > > In terms of likelihood shown below, they are pretty consistent. > > nj=2 -109.265 -> -95.5657 > nj=4 -109.265 -> -95.5599 > nj=8 -109.265 -> -95.556 > > Hao > > On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: >> OK, I have an idea what might have happened. It's possible that because you >> are using such a small number of parallel jobs (2), there is a problem due >> to numerical roundoff. This should not be the case because we accumulate in >> double, but to see whether this is a factor, could you please run the >> training with more jobs? E.g. 4 or 8? See if the likelihoods change. Also, >> show me the full log from the last update, both before and after you make >> the change. >> Dan >> >> >> >> On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User Communication and >> Updates <kal...@li...> wrote: >>> >>> They are almost identical. >>> >>> 011c0201 [ >>> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 >>> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 >>> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 -12.17628 >>> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 >>> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 -3.351704 >>> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 >>> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >>> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >>> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 -9.91317 >>> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 >>> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 >>> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >>> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 -3.151051 >>> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 >>> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 >>> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 >>> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 -10.80855 >>> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 >>> >>> Hao >>> >>> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User >>> Communication and Updates <kal...@li...> wrote: >>> > OK, everything there looks pretty normal. Let's check the features. >>> > Try >>> > doing >>> > >>> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head >>> > copy-feats scp:data/train_si284/feats.scp ark,t:- >>> > 011c0201 [ >>> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 >>> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 >>> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 -12.17628 >>> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 >>> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 -3.351702 >>> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 >>> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >>> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >>> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 >>> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 >>> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 >>> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >>> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 -3.151051 >>> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 >>> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 >>> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 >>> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 -10.80855 >>> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 >>> > >>> > and show me the otuput >>> > Dan >>> > >>> > >>> > >>> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User Communication >>> > and >>> > Updates <kal...@li...> wrote: >>> >> >>> >> Here is the result for >>> >> >>> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >>> >> exp/mono0a/decode/log >>> >> >>> >> Hao >>> >> >>> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User >>> >> Communication and Updates <kal...@li...> wrote: >>> >> > My corresponding numbers are -108.535 -> -95.116, and the difference >>> >> > is >>> >> > about the same. [It's strange that the absolute numbers differ >>> >> > slightly, >>> >> > though.] >>> >> > Could you do >>> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >>> >> > exp/mono0a/decode/log >>> >> > and send the output? >>> >> > >>> >> > Dan >>> >> > >>> >> > >>> >> > >>> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User >>> >> > Communication >>> >> > and >>> >> > Updates <kal...@li...> wrote: >>> >> >> >>> >> >> Copied from exp/mono0a/log/update.39.log >>> >> >> >>> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, >>> >> >> total >>> >> >> count 849568, avg like/frame -95.5657 >>> >> >> >>> >> >> Hao >>> >> >> >>> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User >>> >> >> Communication and Updates <kal...@li...> wrote: >>> >> >> > OK, that is quite weird. What is the avg like/frame on the final >>> >> >> > update.??.log? >>> >> >> > Dan >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User >>> >> >> > Communication >>> >> >> > and >>> >> >> > Updates <kal...@li...> wrote: >>> >> >> >> >>> >> >> >> Unfortunately those are absolute differences. Here is what I get >>> >> >> >> for >>> >> >> >> the >>> >> >> >> mono0a. >>> >> >> >> >>> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 >>> >> >> >> >>> >> >> >> And below is the reported one in RESULTS. >>> >> >> >> >>> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / >>> >> >> >> 5643, >>> >> >> >> 149 >>> >> >> >> ins, 191 del, 1101 sub ] >>> >> >> >> >>> >> >> >> I will try to checkout the same version as the RESULTS file and >>> >> >> >> run >>> >> >> >> the experiments again. >>> >> >> >> >>> >> >> >> Here is the total data count from exp/mono0a/log/update.0.log. >>> >> >> >> >>> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, >>> >> >> >> total >>> >> >> >> count 849568, avg like/frame -109.265 >>> >> >> >> >>> >> >> >> Thanks for your help. >>> >> >> >> >>> >> >> >> Hao >>> >> >> >> >>> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User >>> >> >> >> Communication and Updates <kal...@li...> >>> >> >> >> wrote: >>> >> >> >> > Assuming those are relative numbers, they're probably within >>> >> >> >> > the >>> >> >> >> > margin >>> >> >> >> > of >>> >> >> >> > error. See if it's the same on other test sets. Because >>> >> >> >> > various >>> >> >> >> > algorithms >>> >> >> >> > call rand(), and different machines implement this differently, >>> >> >> >> > results >>> >> >> >> > aren't fully reproducible. >>> >> >> >> > >>> >> >> >> > But to check that it's not some code change that hurts the >>> >> >> >> > results, >>> >> >> >> > you >>> >> >> >> > could try checking out a copy with the same revision number as >>> >> >> >> > the >>> >> >> >> > RESULTS >>> >> >> >> > file, and running again and seeing what results you get. If >>> >> >> >> > there >>> >> >> >> > is >>> >> >> >> > a >>> >> >> >> > difference, I'd like to know. >>> >> >> >> > >>> >> >> >> > It's also possible some files were missing in your WSJ >>> >> >> >> > distribution-- >>> >> >> >> > let me >>> >> >> >> > know the data count reported in one the */log/update.log files >>> >> >> >> > and >>> >> >> >> > I'll >>> >> >> >> > compare with a local copy. >>> >> >> >> > >>> >> >> >> > Dan >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User >>> >> >> >> > Communication >>> >> >> >> > and >>> >> >> >> > Updates <kal...@li...> wrote: >>> >> >> >> >> >>> >> >> >> >> Hi, >>> >> >> >> >> >>> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to >>> >> >> >> >> reproduce >>> >> >> >> >> the >>> >> >> >> >> results stated in the RESULTS file. Are we supposed to get >>> >> >> >> >> exactly >>> >> >> >> >> the >>> >> >> >> >> same numbers with the (hyper-)parameters given in the recipe? >>> >> >> >> >> >>> >> >> >> >> Just to give a sense of how far I'm off. My monophone models >>> >> >> >> >> perform >>> >> >> >> >> 10% >>> >> >> >> >> worse than the reported results on eval92 and the resulting >>> >> >> >> >> triphone >>> >> >> >> >> models perform 4% worse. >>> >> >> >> >> >>> >> >> >> >> Thanks, >>> >> >> >> >> Hao >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> >>> >> >> >> >> ------------------------------------------------------------------------------ >>> >> >> >> >> How ServiceNow helps IT people transform IT departments: >>> >> >> >> >> 1. A cloud service to automate IT design, transition and >>> >> >> >> >> operations >>> >> >> >> >> 2. Dashboards that offer high-level views of enterprise >>> >> >> >> >> services >>> >> >> >> >> 3. A single system of record for all IT processes >>> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >>> >> >> >> >> _______________________________________________ >>> >> >> >> >> Kaldi-users mailing list >>> >> >> >> >> Kal...@li... >>> >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > >>> >> >> >> > ------------------------------------------------------------------------------ >>> >> >> >> > How ServiceNow helps IT people transform IT departments: >>> >> >> >> > 1. A cloud service to automate IT design, transition and >>> >> >> >> > operations >>> >> >> >> > 2. Dashboards that offer high-level views of enterprise >>> >> >> >> > services >>> >> >> >> > 3. A single system of record for all IT processes >>> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >>> >> >> >> > _______________________________________________ >>> >> >> >> > Kaldi-users mailing list >>> >> >> >> > Kal...@li... >>> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> >> >> > >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> >>> >> >> >> ------------------------------------------------------------------------------ >>> >> >> >> How ServiceNow helps IT people transform IT departments: >>> >> >> >> 1. A cloud service to automate IT design, transition and >>> >> >> >> operations >>> >> >> >> 2. Dashboards that offer high-level views of enterprise services >>> >> >> >> 3. A single system of record for all IT processes >>> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >>> >> >> >> _______________________________________________ >>> >> >> >> Kaldi-users mailing list >>> >> >> >> Kal...@li... >>> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > >>> >> >> > ------------------------------------------------------------------------------ >>> >> >> > How ServiceNow helps IT people transform IT departments: >>> >> >> > 1. A cloud service to automate IT design, transition and >>> >> >> > operations >>> >> >> > 2. Dashboards that offer high-level views of enterprise services >>> >> >> > 3. A single system of record for all IT processes >>> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >>> >> >> > _______________________________________________ >>> >> >> > Kaldi-users mailing list >>> >> >> > Kal...@li... >>> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> >> > >>> >> >> >>> >> >> >>> >> >> >>> >> >> >>> >> >> ------------------------------------------------------------------------------ >>> >> >> How ServiceNow helps IT people transform IT departments: >>> >> >> 1. A cloud service to automate IT design, transition and operations >>> >> >> 2. Dashboards that offer high-level views of enterprise services >>> >> >> 3. A single system of record for all IT processes >>> >> >> http://p.sf.net/sfu/servicenow-d2d-j >>> >> >> _______________________________________________ >>> >> >> Kaldi-users mailing list >>> >> >> Kal...@li... >>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> > >>> >> > >>> >> > >>> >> > >>> >> > >>> >> > ------------------------------------------------------------------------------ >>> >> > How ServiceNow helps IT people transform IT departments: >>> >> > 1. A cloud service to automate IT design, transition and operations >>> >> > 2. Dashboards that offer high-level views of enterprise services >>> >> > 3. A single system of record for all IT processes >>> >> > http://p.sf.net/sfu/servicenow-d2d-j >>> >> > _______________________________________________ >>> >> > Kaldi-users mailing list >>> >> > Kal...@li... >>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> > >>> >> >>> >> >>> >> >>> >> ------------------------------------------------------------------------------ >>> >> How ServiceNow helps IT people transform IT departments: >>> >> 1. A cloud service to automate IT design, transition and operations >>> >> 2. Dashboards that offer high-level views of enterprise services >>> >> 3. A single system of record for all IT processes >>> >> http://p.sf.net/sfu/servicenow-d2d-j >>> >> _______________________________________________ >>> >> Kaldi-users mailing list >>> >> Kal...@li... >>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >> >>> > >>> > >>> > >>> > ------------------------------------------------------------------------------ >>> > How ServiceNow helps IT people transform IT departments: >>> > 1. A cloud service to automate IT design, transition and operations >>> > 2. Dashboards that offer high-level views of enterprise services >>> > 3. A single system of record for all IT processes >>> > http://p.sf.net/sfu/servicenow-d2d-j >>> > _______________________________________________ >>> > Kaldi-users mailing list >>> > Kal...@li... >>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> > >>> >>> >>> ------------------------------------------------------------------------------ >>> How ServiceNow helps IT people transform IT departments: >>> 1. A cloud service to automate IT design, transition and operations >>> 2. Dashboards that offer high-level views of enterprise services >>> 3. A single system of record for all IT processes >>> http://p.sf.net/sfu/servicenow-d2d-j >>> _______________________________________________ >>> Kaldi-users mailing list >>> Kal...@li... >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> >> ------------------------------------------------------------------------------ >> How ServiceNow helps IT people transform IT departments: >> 1. A cloud service to automate IT design, transition and operations >> 2. Dashboards that offer high-level views of enterprise services >> 3. A single system of record for all IT processes >> http://p.sf.net/sfu/servicenow-d2d-j >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-04 07:02:55
|
Here are the log for the last update with different number of jobs. I don't see much of a difference. In terms of likelihood shown below, they are pretty consistent. nj=2 -109.265 -> -95.5657 nj=4 -109.265 -> -95.5599 nj=8 -109.265 -> -95.556 Hao On Mon, Jun 3, 2013 at 5:29 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > OK, I have an idea what might have happened. It's possible that because you > are using such a small number of parallel jobs (2), there is a problem due > to numerical roundoff. This should not be the case because we accumulate in > double, but to see whether this is a factor, could you please run the > training with more jobs? E.g. 4 or 8? See if the likelihoods change. Also, > show me the full log from the last update, both before and after you make > the change. > Dan > > > > On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User Communication and > Updates <kal...@li...> wrote: >> >> They are almost identical. >> >> 011c0201 [ >> 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 >> 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 >> 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 -12.17628 >> -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 >> 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 -3.351704 >> -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 >> 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >> -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >> 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 -9.91317 >> -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 >> 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 >> -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >> 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 -3.151051 >> 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 >> 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 >> 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 >> 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 -10.80855 >> -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 >> >> Hao >> >> On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User >> Communication and Updates <kal...@li...> wrote: >> > OK, everything there looks pretty normal. Let's check the features. >> > Try >> > doing >> > >> > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head >> > copy-feats scp:data/train_si284/feats.scp ark,t:- >> > 011c0201 [ >> > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 >> > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 >> > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 -12.17628 >> > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 >> > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 -3.351702 >> > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 >> > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 >> > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 >> > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 >> > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 >> > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 >> > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 >> > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 -3.151051 >> > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 >> > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 >> > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 >> > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 -10.80855 >> > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 >> > >> > and show me the otuput >> > Dan >> > >> > >> > >> > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User Communication >> > and >> > Updates <kal...@li...> wrote: >> >> >> >> Here is the result for >> >> >> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> >> exp/mono0a/decode/log >> >> >> >> Hao >> >> >> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User >> >> Communication and Updates <kal...@li...> wrote: >> >> > My corresponding numbers are -108.535 -> -95.116, and the difference >> >> > is >> >> > about the same. [It's strange that the absolute numbers differ >> >> > slightly, >> >> > though.] >> >> > Could you do >> >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> >> > exp/mono0a/decode/log >> >> > and send the output? >> >> > >> >> > Dan >> >> > >> >> > >> >> > >> >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User >> >> > Communication >> >> > and >> >> > Updates <kal...@li...> wrote: >> >> >> >> >> >> Copied from exp/mono0a/log/update.39.log >> >> >> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, >> >> >> total >> >> >> count 849568, avg like/frame -95.5657 >> >> >> >> >> >> Hao >> >> >> >> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User >> >> >> Communication and Updates <kal...@li...> wrote: >> >> >> > OK, that is quite weird. What is the avg like/frame on the final >> >> >> > update.??.log? >> >> >> > Dan >> >> >> > >> >> >> > >> >> >> > >> >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User >> >> >> > Communication >> >> >> > and >> >> >> > Updates <kal...@li...> wrote: >> >> >> >> >> >> >> >> Unfortunately those are absolute differences. Here is what I get >> >> >> >> for >> >> >> >> the >> >> >> >> mono0a. >> >> >> >> >> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] >> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 >> >> >> >> >> >> >> >> And below is the reported one in RESULTS. >> >> >> >> >> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / >> >> >> >> 5643, >> >> >> >> 149 >> >> >> >> ins, 191 del, 1101 sub ] >> >> >> >> >> >> >> >> I will try to checkout the same version as the RESULTS file and >> >> >> >> run >> >> >> >> the experiments again. >> >> >> >> >> >> >> >> Here is the total data count from exp/mono0a/log/update.0.log. >> >> >> >> >> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, >> >> >> >> total >> >> >> >> count 849568, avg like/frame -109.265 >> >> >> >> >> >> >> >> Thanks for your help. >> >> >> >> >> >> >> >> Hao >> >> >> >> >> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User >> >> >> >> Communication and Updates <kal...@li...> >> >> >> >> wrote: >> >> >> >> > Assuming those are relative numbers, they're probably within >> >> >> >> > the >> >> >> >> > margin >> >> >> >> > of >> >> >> >> > error. See if it's the same on other test sets. Because >> >> >> >> > various >> >> >> >> > algorithms >> >> >> >> > call rand(), and different machines implement this differently, >> >> >> >> > results >> >> >> >> > aren't fully reproducible. >> >> >> >> > >> >> >> >> > But to check that it's not some code change that hurts the >> >> >> >> > results, >> >> >> >> > you >> >> >> >> > could try checking out a copy with the same revision number as >> >> >> >> > the >> >> >> >> > RESULTS >> >> >> >> > file, and running again and seeing what results you get. If >> >> >> >> > there >> >> >> >> > is >> >> >> >> > a >> >> >> >> > difference, I'd like to know. >> >> >> >> > >> >> >> >> > It's also possible some files were missing in your WSJ >> >> >> >> > distribution-- >> >> >> >> > let me >> >> >> >> > know the data count reported in one the */log/update.log files >> >> >> >> > and >> >> >> >> > I'll >> >> >> >> > compare with a local copy. >> >> >> >> > >> >> >> >> > Dan >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User >> >> >> >> > Communication >> >> >> >> > and >> >> >> >> > Updates <kal...@li...> wrote: >> >> >> >> >> >> >> >> >> >> Hi, >> >> >> >> >> >> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to >> >> >> >> >> reproduce >> >> >> >> >> the >> >> >> >> >> results stated in the RESULTS file. Are we supposed to get >> >> >> >> >> exactly >> >> >> >> >> the >> >> >> >> >> same numbers with the (hyper-)parameters given in the recipe? >> >> >> >> >> >> >> >> >> >> Just to give a sense of how far I'm off. My monophone models >> >> >> >> >> perform >> >> >> >> >> 10% >> >> >> >> >> worse than the reported results on eval92 and the resulting >> >> >> >> >> triphone >> >> >> >> >> models perform 4% worse. >> >> >> >> >> >> >> >> >> >> Thanks, >> >> >> >> >> Hao >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >> >> >> 1. A cloud service to automate IT design, transition and >> >> >> >> >> operations >> >> >> >> >> 2. Dashboards that offer high-level views of enterprise >> >> >> >> >> services >> >> >> >> >> 3. A single system of record for all IT processes >> >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >> >> >> _______________________________________________ >> >> >> >> >> Kaldi-users mailing list >> >> >> >> >> Kal...@li... >> >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > ------------------------------------------------------------------------------ >> >> >> >> > How ServiceNow helps IT people transform IT departments: >> >> >> >> > 1. A cloud service to automate IT design, transition and >> >> >> >> > operations >> >> >> >> > 2. Dashboards that offer high-level views of enterprise >> >> >> >> > services >> >> >> >> > 3. A single system of record for all IT processes >> >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >> >> > _______________________________________________ >> >> >> >> > Kaldi-users mailing list >> >> >> >> > Kal...@li... >> >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> >> > >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >> >> 1. A cloud service to automate IT design, transition and >> >> >> >> operations >> >> >> >> 2. Dashboards that offer high-level views of enterprise services >> >> >> >> 3. A single system of record for all IT processes >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >> >> _______________________________________________ >> >> >> >> Kaldi-users mailing list >> >> >> >> Kal...@li... >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > ------------------------------------------------------------------------------ >> >> >> > How ServiceNow helps IT people transform IT departments: >> >> >> > 1. A cloud service to automate IT design, transition and >> >> >> > operations >> >> >> > 2. Dashboards that offer high-level views of enterprise services >> >> >> > 3. A single system of record for all IT processes >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >> > _______________________________________________ >> >> >> > Kaldi-users mailing list >> >> >> > Kal...@li... >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> > >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >> 1. A cloud service to automate IT design, transition and operations >> >> >> 2. Dashboards that offer high-level views of enterprise services >> >> >> 3. A single system of record for all IT processes >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >> _______________________________________________ >> >> >> Kaldi-users mailing list >> >> >> Kal...@li... >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ >> >> > How ServiceNow helps IT people transform IT departments: >> >> > 1. A cloud service to automate IT design, transition and operations >> >> > 2. Dashboards that offer high-level views of enterprise services >> >> > 3. A single system of record for all IT processes >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> > _______________________________________________ >> >> > Kaldi-users mailing list >> >> > Kal...@li... >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> > >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> How ServiceNow helps IT people transform IT departments: >> >> 1. A cloud service to automate IT design, transition and operations >> >> 2. Dashboards that offer high-level views of enterprise services >> >> 3. A single system of record for all IT processes >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> _______________________________________________ >> >> Kaldi-users mailing list >> >> Kal...@li... >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> > >> > >> > >> > ------------------------------------------------------------------------------ >> > How ServiceNow helps IT people transform IT departments: >> > 1. A cloud service to automate IT design, transition and operations >> > 2. Dashboards that offer high-level views of enterprise services >> > 3. A single system of record for all IT processes >> > http://p.sf.net/sfu/servicenow-d2d-j >> > _______________________________________________ >> > Kaldi-users mailing list >> > Kal...@li... >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > >> >> >> ------------------------------------------------------------------------------ >> How ServiceNow helps IT people transform IT departments: >> 1. A cloud service to automate IT design, transition and operations >> 2. Dashboards that offer high-level views of enterprise services >> 3. A single system of record for all IT processes >> http://p.sf.net/sfu/servicenow-d2d-j >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 22:29:34
|
OK, I have an idea what might have happened. It's possible that because you are using such a small number of parallel jobs (2), there is a problem due to numerical roundoff. This should not be the case because we accumulate in double, but to see whether this is a factor, could you please run the training with more jobs? E.g. 4 or 8? See if the likelihoods change. Also, show me the full log from the last update, both before and after you make the change. Dan On Mon, Jun 3, 2013 at 6:23 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > They are almost identical. > > 011c0201 [ > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 > 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 -12.17628 > -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 > 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 -3.351704 > -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 -9.91317 > -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 -3.151051 > 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 > 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 > 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 > 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 -10.80855 > -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 > > Hao > > On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > OK, everything there looks pretty normal. Let's check the features. Try > > doing > > > > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head > > copy-feats scp:data/train_si284/feats.scp ark,t:- > > 011c0201 [ > > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 > > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 > > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 -12.17628 > > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 > > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 -3.351702 > > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 > > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 > > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 > > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 > > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 > > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 -3.151051 > > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 > > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 > > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 > > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 -10.80855 > > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 > > > > and show me the otuput > > Dan > > > > > > > > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User Communication > and > > Updates <kal...@li...> wrote: > >> > >> Here is the result for > >> > >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >> exp/mono0a/decode/log > >> > >> Hao > >> > >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User > >> Communication and Updates <kal...@li...> wrote: > >> > My corresponding numbers are -108.535 -> -95.116, and the difference > is > >> > about the same. [It's strange that the absolute numbers differ > >> > slightly, > >> > though.] > >> > Could you do > >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > >> > exp/mono0a/decode/log > >> > and send the output? > >> > > >> > Dan > >> > > >> > > >> > > >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User > Communication > >> > and > >> > Updates <kal...@li...> wrote: > >> >> > >> >> Copied from exp/mono0a/log/update.39.log > >> >> > >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, > total > >> >> count 849568, avg like/frame -95.5657 > >> >> > >> >> Hao > >> >> > >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User > >> >> Communication and Updates <kal...@li...> wrote: > >> >> > OK, that is quite weird. What is the avg like/frame on the final > >> >> > update.??.log? > >> >> > Dan > >> >> > > >> >> > > >> >> > > >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User > >> >> > Communication > >> >> > and > >> >> > Updates <kal...@li...> wrote: > >> >> >> > >> >> >> Unfortunately those are absolute differences. Here is what I get > for > >> >> >> the > >> >> >> mono0a. > >> >> >> > >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] > >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 > >> >> >> > >> >> >> And below is the reported one in RESULTS. > >> >> >> > >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, > >> >> >> 149 > >> >> >> ins, 191 del, 1101 sub ] > >> >> >> > >> >> >> I will try to checkout the same version as the RESULTS file and > run > >> >> >> the experiments again. > >> >> >> > >> >> >> Here is the total data count from exp/mono0a/log/update.0.log. > >> >> >> > >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, > >> >> >> total > >> >> >> count 849568, avg like/frame -109.265 > >> >> >> > >> >> >> Thanks for your help. > >> >> >> > >> >> >> Hao > >> >> >> > >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User > >> >> >> Communication and Updates <kal...@li...> > wrote: > >> >> >> > Assuming those are relative numbers, they're probably within the > >> >> >> > margin > >> >> >> > of > >> >> >> > error. See if it's the same on other test sets. Because > various > >> >> >> > algorithms > >> >> >> > call rand(), and different machines implement this differently, > >> >> >> > results > >> >> >> > aren't fully reproducible. > >> >> >> > > >> >> >> > But to check that it's not some code change that hurts the > >> >> >> > results, > >> >> >> > you > >> >> >> > could try checking out a copy with the same revision number as > the > >> >> >> > RESULTS > >> >> >> > file, and running again and seeing what results you get. If > there > >> >> >> > is > >> >> >> > a > >> >> >> > difference, I'd like to know. > >> >> >> > > >> >> >> > It's also possible some files were missing in your WSJ > >> >> >> > distribution-- > >> >> >> > let me > >> >> >> > know the data count reported in one the */log/update.log files > and > >> >> >> > I'll > >> >> >> > compare with a local copy. > >> >> >> > > >> >> >> > Dan > >> >> >> > > >> >> >> > > >> >> >> > > >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User > >> >> >> > Communication > >> >> >> > and > >> >> >> > Updates <kal...@li...> wrote: > >> >> >> >> > >> >> >> >> Hi, > >> >> >> >> > >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce > >> >> >> >> the > >> >> >> >> results stated in the RESULTS file. Are we supposed to get > >> >> >> >> exactly > >> >> >> >> the > >> >> >> >> same numbers with the (hyper-)parameters given in the recipe? > >> >> >> >> > >> >> >> >> Just to give a sense of how far I'm off. My monophone models > >> >> >> >> perform > >> >> >> >> 10% > >> >> >> >> worse than the reported results on eval92 and the resulting > >> >> >> >> triphone > >> >> >> >> models perform 4% worse. > >> >> >> >> > >> >> >> >> Thanks, > >> >> >> >> Hao > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > >> >> >> >> > ------------------------------------------------------------------------------ > >> >> >> >> How ServiceNow helps IT people transform IT departments: > >> >> >> >> 1. A cloud service to automate IT design, transition and > >> >> >> >> operations > >> >> >> >> 2. Dashboards that offer high-level views of enterprise > services > >> >> >> >> 3. A single system of record for all IT processes > >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >> >> _______________________________________________ > >> >> >> >> Kaldi-users mailing list > >> >> >> >> Kal...@li... > >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >> > > >> >> >> > > >> >> >> > > >> >> >> > > >> >> >> > > >> >> >> > > >> >> >> > > ------------------------------------------------------------------------------ > >> >> >> > How ServiceNow helps IT people transform IT departments: > >> >> >> > 1. A cloud service to automate IT design, transition and > >> >> >> > operations > >> >> >> > 2. Dashboards that offer high-level views of enterprise services > >> >> >> > 3. A single system of record for all IT processes > >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> >> > _______________________________________________ > >> >> >> > Kaldi-users mailing list > >> >> >> > Kal...@li... > >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> >> > > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > ------------------------------------------------------------------------------ > >> >> >> How ServiceNow helps IT people transform IT departments: > >> >> >> 1. A cloud service to automate IT design, transition and > operations > >> >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> >> 3. A single system of record for all IT processes > >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >> _______________________________________________ > >> >> >> Kaldi-users mailing list > >> >> >> Kal...@li... > >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > > >> >> > > >> >> > > >> >> > > >> >> > > >> >> > > ------------------------------------------------------------------------------ > >> >> > How ServiceNow helps IT people transform IT departments: > >> >> > 1. A cloud service to automate IT design, transition and operations > >> >> > 2. Dashboards that offer high-level views of enterprise services > >> >> > 3. A single system of record for all IT processes > >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> > _______________________________________________ > >> >> > Kaldi-users mailing list > >> >> > Kal...@li... > >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ > >> >> How ServiceNow helps IT people transform IT departments: > >> >> 1. A cloud service to automate IT design, transition and operations > >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> 3. A single system of record for all IT processes > >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> _______________________________________________ > >> >> Kaldi-users mailing list > >> >> Kal...@li... > >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > > >> > > >> > > >> > > ------------------------------------------------------------------------------ > >> > How ServiceNow helps IT people transform IT departments: > >> > 1. A cloud service to automate IT design, transition and operations > >> > 2. Dashboards that offer high-level views of enterprise services > >> > 3. A single system of record for all IT processes > >> > http://p.sf.net/sfu/servicenow-d2d-j > >> > _______________________________________________ > >> > Kaldi-users mailing list > >> > Kal...@li... > >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 22:23:54
|
They are almost identical. 011c0201 [ 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 2.553219 3.935668 10.76861 9.013732 4.780495 -3.26797 46.34384 -33.52037 -10.28797 -11.47937 -5.18444 -6.867837 -12.17628 -6.932252 -7.423642 -1.106863 3.337547 -0.5728683 0.6971088 46.81014 -31.05773 -7.440363 -11.46406 -5.926405 -2.568082 -3.351704 -5.60004 -8.281031 -9.46069 6.799883 6.513657 -7.97058 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 46.92208 -33.49314 -9.207784 -12.05287 -5.494797 -4.269251 -9.91317 -1.722169 -0.7680084 -2.61649 9.212611 7.324253 12.41928 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 47.34614 -33.04692 -7.571927 -8.986547 -3.523783 1.130056 -3.151051 3.572649 3.381185 -3.117306 -2.670765 2.165989 0.3512498 46.63126 -33.57331 -11.13035 -7.506401 2.113542 3.268812 -1.814186 2.770102 -4.381135 -3.264607 -3.549407 2.463478 -0.7675163 46.31366 -33.00954 -6.964884 -6.109343 -7.121595 -4.326707 -10.80855 -6.96949 6.597637 4.109591 1.001096 2.592361 6.154362 Hao On Mon, Jun 3, 2013 at 5:11 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > OK, everything there looks pretty normal. Let's check the features. Try > doing > > path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head > copy-feats scp:data/train_si284/feats.scp ark,t:- > 011c0201 [ > 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 > 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 > 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 -12.17628 > -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 > 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 -3.351702 > -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 > 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 > -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 > 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 > -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 > 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 > -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 > 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 -3.151051 > 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 > 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 > 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 > 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 -10.80855 > -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 > > and show me the otuput > Dan > > > > On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User Communication and > Updates <kal...@li...> wrote: >> >> Here is the result for >> >> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> exp/mono0a/decode/log >> >> Hao >> >> On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User >> Communication and Updates <kal...@li...> wrote: >> > My corresponding numbers are -108.535 -> -95.116, and the difference is >> > about the same. [It's strange that the absolute numbers differ >> > slightly, >> > though.] >> > Could you do >> > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log >> > exp/mono0a/decode/log >> > and send the output? >> > >> > Dan >> > >> > >> > >> > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User Communication >> > and >> > Updates <kal...@li...> wrote: >> >> >> >> Copied from exp/mono0a/log/update.39.log >> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total >> >> count 849568, avg like/frame -95.5657 >> >> >> >> Hao >> >> >> >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User >> >> Communication and Updates <kal...@li...> wrote: >> >> > OK, that is quite weird. What is the avg like/frame on the final >> >> > update.??.log? >> >> > Dan >> >> > >> >> > >> >> > >> >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User >> >> > Communication >> >> > and >> >> > Updates <kal...@li...> wrote: >> >> >> >> >> >> Unfortunately those are absolute differences. Here is what I get for >> >> >> the >> >> >> mono0a. >> >> >> >> >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] >> >> >> exp/mono0a/decode_tgpr_eval92/wer_9 >> >> >> >> >> >> And below is the reported one in RESULTS. >> >> >> >> >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, >> >> >> 149 >> >> >> ins, 191 del, 1101 sub ] >> >> >> >> >> >> I will try to checkout the same version as the RESULTS file and run >> >> >> the experiments again. >> >> >> >> >> >> Here is the total data count from exp/mono0a/log/update.0.log. >> >> >> >> >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, >> >> >> total >> >> >> count 849568, avg like/frame -109.265 >> >> >> >> >> >> Thanks for your help. >> >> >> >> >> >> Hao >> >> >> >> >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User >> >> >> Communication and Updates <kal...@li...> wrote: >> >> >> > Assuming those are relative numbers, they're probably within the >> >> >> > margin >> >> >> > of >> >> >> > error. See if it's the same on other test sets. Because various >> >> >> > algorithms >> >> >> > call rand(), and different machines implement this differently, >> >> >> > results >> >> >> > aren't fully reproducible. >> >> >> > >> >> >> > But to check that it's not some code change that hurts the >> >> >> > results, >> >> >> > you >> >> >> > could try checking out a copy with the same revision number as the >> >> >> > RESULTS >> >> >> > file, and running again and seeing what results you get. If there >> >> >> > is >> >> >> > a >> >> >> > difference, I'd like to know. >> >> >> > >> >> >> > It's also possible some files were missing in your WSJ >> >> >> > distribution-- >> >> >> > let me >> >> >> > know the data count reported in one the */log/update.log files and >> >> >> > I'll >> >> >> > compare with a local copy. >> >> >> > >> >> >> > Dan >> >> >> > >> >> >> > >> >> >> > >> >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User >> >> >> > Communication >> >> >> > and >> >> >> > Updates <kal...@li...> wrote: >> >> >> >> >> >> >> >> Hi, >> >> >> >> >> >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce >> >> >> >> the >> >> >> >> results stated in the RESULTS file. Are we supposed to get >> >> >> >> exactly >> >> >> >> the >> >> >> >> same numbers with the (hyper-)parameters given in the recipe? >> >> >> >> >> >> >> >> Just to give a sense of how far I'm off. My monophone models >> >> >> >> perform >> >> >> >> 10% >> >> >> >> worse than the reported results on eval92 and the resulting >> >> >> >> triphone >> >> >> >> models perform 4% worse. >> >> >> >> >> >> >> >> Thanks, >> >> >> >> Hao >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >> >> 1. A cloud service to automate IT design, transition and >> >> >> >> operations >> >> >> >> 2. Dashboards that offer high-level views of enterprise services >> >> >> >> 3. A single system of record for all IT processes >> >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >> >> _______________________________________________ >> >> >> >> Kaldi-users mailing list >> >> >> >> Kal...@li... >> >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > ------------------------------------------------------------------------------ >> >> >> > How ServiceNow helps IT people transform IT departments: >> >> >> > 1. A cloud service to automate IT design, transition and >> >> >> > operations >> >> >> > 2. Dashboards that offer high-level views of enterprise services >> >> >> > 3. A single system of record for all IT processes >> >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> >> > _______________________________________________ >> >> >> > Kaldi-users mailing list >> >> >> > Kal...@li... >> >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> >> > >> >> >> >> >> >> >> >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> >> How ServiceNow helps IT people transform IT departments: >> >> >> 1. A cloud service to automate IT design, transition and operations >> >> >> 2. Dashboards that offer high-level views of enterprise services >> >> >> 3. A single system of record for all IT processes >> >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> >> _______________________________________________ >> >> >> Kaldi-users mailing list >> >> >> Kal...@li... >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ >> >> > How ServiceNow helps IT people transform IT departments: >> >> > 1. A cloud service to automate IT design, transition and operations >> >> > 2. Dashboards that offer high-level views of enterprise services >> >> > 3. A single system of record for all IT processes >> >> > http://p.sf.net/sfu/servicenow-d2d-j >> >> > _______________________________________________ >> >> > Kaldi-users mailing list >> >> > Kal...@li... >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> > >> >> >> >> >> >> >> >> ------------------------------------------------------------------------------ >> >> How ServiceNow helps IT people transform IT departments: >> >> 1. A cloud service to automate IT design, transition and operations >> >> 2. Dashboards that offer high-level views of enterprise services >> >> 3. A single system of record for all IT processes >> >> http://p.sf.net/sfu/servicenow-d2d-j >> >> _______________________________________________ >> >> Kaldi-users mailing list >> >> Kal...@li... >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > >> > >> > >> > >> > ------------------------------------------------------------------------------ >> > How ServiceNow helps IT people transform IT departments: >> > 1. A cloud service to automate IT design, transition and operations >> > 2. Dashboards that offer high-level views of enterprise services >> > 3. A single system of record for all IT processes >> > http://p.sf.net/sfu/servicenow-d2d-j >> > _______________________________________________ >> > Kaldi-users mailing list >> > Kal...@li... >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > >> >> >> ------------------------------------------------------------------------------ >> How ServiceNow helps IT people transform IT departments: >> 1. A cloud service to automate IT design, transition and operations >> 2. Dashboards that offer high-level views of enterprise services >> 3. A single system of record for all IT processes >> http://p.sf.net/sfu/servicenow-d2d-j >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 22:12:02
|
OK, everything there looks pretty normal. Let's check the features. Try doing path.sh; copy-feats scp:data/train_si284/feats.scp ark,t:- | head copy-feats scp:data/train_si284/feats.scp ark,t:- 011c0201 [ 46.88603 -32.66241 -11.9018 -14.99819 -4.834829 -2.596024 -5.20842 2.553219 3.935668 10.76861 9.013732 4.780495 -3.267961 46.34384 -33.52037 -10.28797 -11.47937 -5.184441 -6.867837 -12.17628 -6.93225 -7.423645 -1.106866 3.337547 -0.5728683 0.6970833 46.81014 -31.05773 -7.440367 -11.46406 -5.926404 -2.568073 -3.351702 -5.60004 -8.281028 -9.460709 6.799894 6.513657 -7.970572 46.8651 -30.11357 -5.052508 -11.06986 -6.155907 -2.664125 -10.907 -8.487434 -3.194182 -6.367419 6.390572 0.01187897 -0.1876552 46.92208 -33.49314 -9.207782 -12.05287 -5.4948 -4.269254 -9.913167 -1.722165 -0.7680123 -2.616493 9.212611 7.324253 12.41928 46.8317 -32.19164 -8.810924 -12.31202 -3.736232 -1.335111 -8.361964 -2.103669 -3.199154 -6.140819 -8.036694 -8.49004 2.942402 47.34614 -33.04692 -7.571927 -8.986545 -3.523783 1.130052 -3.151051 3.572651 3.381188 -3.117314 -2.670754 2.165966 0.3512328 46.63126 -33.57331 -11.13035 -7.506401 2.113541 3.268804 -1.814186 2.770107 -4.381127 -3.264618 -3.549418 2.463501 -0.7675107 46.31366 -33.00954 -6.964884 -6.109342 -7.121596 -4.326709 -10.80855 -6.96948 6.597637 4.109591 1.001096 2.592361 6.154356 and show me the otuput Dan On Mon, Jun 3, 2013 at 5:59 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Here is the result for > > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > exp/mono0a/decode/log > > Hao > > On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > My corresponding numbers are -108.535 -> -95.116, and the difference is > > about the same. [It's strange that the absolute numbers differ slightly, > > though.] > > Could you do > > ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log > > exp/mono0a/decode/log > > and send the output? > > > > Dan > > > > > > > > On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User Communication > and > > Updates <kal...@li...> wrote: > >> > >> Copied from exp/mono0a/log/update.39.log > >> > >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total > >> count 849568, avg like/frame -95.5657 > >> > >> Hao > >> > >> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User > >> Communication and Updates <kal...@li...> wrote: > >> > OK, that is quite weird. What is the avg like/frame on the final > >> > update.??.log? > >> > Dan > >> > > >> > > >> > > >> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User > Communication > >> > and > >> > Updates <kal...@li...> wrote: > >> >> > >> >> Unfortunately those are absolute differences. Here is what I get for > >> >> the > >> >> mono0a. > >> >> > >> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] > >> >> exp/mono0a/decode_tgpr_eval92/wer_9 > >> >> > >> >> And below is the reported one in RESULTS. > >> >> > >> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, > 149 > >> >> ins, 191 del, 1101 sub ] > >> >> > >> >> I will try to checkout the same version as the RESULTS file and run > >> >> the experiments again. > >> >> > >> >> Here is the total data count from exp/mono0a/log/update.0.log. > >> >> > >> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, > total > >> >> count 849568, avg like/frame -109.265 > >> >> > >> >> Thanks for your help. > >> >> > >> >> Hao > >> >> > >> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User > >> >> Communication and Updates <kal...@li...> wrote: > >> >> > Assuming those are relative numbers, they're probably within the > >> >> > margin > >> >> > of > >> >> > error. See if it's the same on other test sets. Because various > >> >> > algorithms > >> >> > call rand(), and different machines implement this differently, > >> >> > results > >> >> > aren't fully reproducible. > >> >> > > >> >> > But to check that it's not some code change that hurts the results, > >> >> > you > >> >> > could try checking out a copy with the same revision number as the > >> >> > RESULTS > >> >> > file, and running again and seeing what results you get. If there > is > >> >> > a > >> >> > difference, I'd like to know. > >> >> > > >> >> > It's also possible some files were missing in your WSJ > distribution-- > >> >> > let me > >> >> > know the data count reported in one the */log/update.log files and > >> >> > I'll > >> >> > compare with a local copy. > >> >> > > >> >> > Dan > >> >> > > >> >> > > >> >> > > >> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User > >> >> > Communication > >> >> > and > >> >> > Updates <kal...@li...> wrote: > >> >> >> > >> >> >> Hi, > >> >> >> > >> >> >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce > the > >> >> >> results stated in the RESULTS file. Are we supposed to get exactly > >> >> >> the > >> >> >> same numbers with the (hyper-)parameters given in the recipe? > >> >> >> > >> >> >> Just to give a sense of how far I'm off. My monophone models > perform > >> >> >> 10% > >> >> >> worse than the reported results on eval92 and the resulting > triphone > >> >> >> models perform 4% worse. > >> >> >> > >> >> >> Thanks, > >> >> >> Hao > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > ------------------------------------------------------------------------------ > >> >> >> How ServiceNow helps IT people transform IT departments: > >> >> >> 1. A cloud service to automate IT design, transition and > operations > >> >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> >> 3. A single system of record for all IT processes > >> >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> >> _______________________________________________ > >> >> >> Kaldi-users mailing list > >> >> >> Kal...@li... > >> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > > >> >> > > >> >> > > >> >> > > >> >> > > >> >> > > ------------------------------------------------------------------------------ > >> >> > How ServiceNow helps IT people transform IT departments: > >> >> > 1. A cloud service to automate IT design, transition and operations > >> >> > 2. Dashboards that offer high-level views of enterprise services > >> >> > 3. A single system of record for all IT processes > >> >> > http://p.sf.net/sfu/servicenow-d2d-j > >> >> > _______________________________________________ > >> >> > Kaldi-users mailing list > >> >> > Kal...@li... > >> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> >> > > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ > >> >> How ServiceNow helps IT people transform IT departments: > >> >> 1. A cloud service to automate IT design, transition and operations > >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> 3. A single system of record for all IT processes > >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> _______________________________________________ > >> >> Kaldi-users mailing list > >> >> Kal...@li... > >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > > >> > > >> > > >> > > ------------------------------------------------------------------------------ > >> > How ServiceNow helps IT people transform IT departments: > >> > 1. A cloud service to automate IT design, transition and operations > >> > 2. Dashboards that offer high-level views of enterprise services > >> > 3. A single system of record for all IT processes > >> > http://p.sf.net/sfu/servicenow-d2d-j > >> > _______________________________________________ > >> > Kaldi-users mailing list > >> > Kal...@li... > >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 21:59:47
|
Here is the result for
ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log
exp/mono0a/decode/log
Hao
On Mon, Jun 3, 2013 at 4:49 PM, Mailing list used for User
Communication and Updates <kal...@li...> wrote:
> My corresponding numbers are -108.535 -> -95.116, and the difference is
> about the same. [It's strange that the absolute numbers differ slightly,
> though.]
> Could you do
> ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log
> exp/mono0a/decode/log
> and send the output?
>
> Dan
>
>
>
> On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User Communication and
> Updates <kal...@li...> wrote:
>>
>> Copied from exp/mono0a/log/update.39.log
>>
>> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total
>> count 849568, avg like/frame -95.5657
>>
>> Hao
>>
>> On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User
>> Communication and Updates <kal...@li...> wrote:
>> > OK, that is quite weird. What is the avg like/frame on the final
>> > update.??.log?
>> > Dan
>> >
>> >
>> >
>> > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User Communication
>> > and
>> > Updates <kal...@li...> wrote:
>> >>
>> >> Unfortunately those are absolute differences. Here is what I get for
>> >> the
>> >> mono0a.
>> >>
>> >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ]
>> >> exp/mono0a/decode_tgpr_eval92/wer_9
>> >>
>> >> And below is the reported one in RESULTS.
>> >>
>> >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, 149
>> >> ins, 191 del, 1101 sub ]
>> >>
>> >> I will try to checkout the same version as the RESULTS file and run
>> >> the experiments again.
>> >>
>> >> Here is the total data count from exp/mono0a/log/update.0.log.
>> >>
>> >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total
>> >> count 849568, avg like/frame -109.265
>> >>
>> >> Thanks for your help.
>> >>
>> >> Hao
>> >>
>> >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User
>> >> Communication and Updates <kal...@li...> wrote:
>> >> > Assuming those are relative numbers, they're probably within the
>> >> > margin
>> >> > of
>> >> > error. See if it's the same on other test sets. Because various
>> >> > algorithms
>> >> > call rand(), and different machines implement this differently,
>> >> > results
>> >> > aren't fully reproducible.
>> >> >
>> >> > But to check that it's not some code change that hurts the results,
>> >> > you
>> >> > could try checking out a copy with the same revision number as the
>> >> > RESULTS
>> >> > file, and running again and seeing what results you get. If there is
>> >> > a
>> >> > difference, I'd like to know.
>> >> >
>> >> > It's also possible some files were missing in your WSJ distribution--
>> >> > let me
>> >> > know the data count reported in one the */log/update.log files and
>> >> > I'll
>> >> > compare with a local copy.
>> >> >
>> >> > Dan
>> >> >
>> >> >
>> >> >
>> >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User
>> >> > Communication
>> >> > and
>> >> > Updates <kal...@li...> wrote:
>> >> >>
>> >> >> Hi,
>> >> >>
>> >> >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce the
>> >> >> results stated in the RESULTS file. Are we supposed to get exactly
>> >> >> the
>> >> >> same numbers with the (hyper-)parameters given in the recipe?
>> >> >>
>> >> >> Just to give a sense of how far I'm off. My monophone models perform
>> >> >> 10%
>> >> >> worse than the reported results on eval92 and the resulting triphone
>> >> >> models perform 4% worse.
>> >> >>
>> >> >> Thanks,
>> >> >> Hao
>> >> >>
>> >> >>
>> >> >>
>> >> >>
>> >> >> ------------------------------------------------------------------------------
>> >> >> How ServiceNow helps IT people transform IT departments:
>> >> >> 1. A cloud service to automate IT design, transition and operations
>> >> >> 2. Dashboards that offer high-level views of enterprise services
>> >> >> 3. A single system of record for all IT processes
>> >> >> http://p.sf.net/sfu/servicenow-d2d-j
>> >> >> _______________________________________________
>> >> >> Kaldi-users mailing list
>> >> >> Kal...@li...
>> >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>> >> >
>> >> >
>> >> >
>> >> >
>> >> >
>> >> > ------------------------------------------------------------------------------
>> >> > How ServiceNow helps IT people transform IT departments:
>> >> > 1. A cloud service to automate IT design, transition and operations
>> >> > 2. Dashboards that offer high-level views of enterprise services
>> >> > 3. A single system of record for all IT processes
>> >> > http://p.sf.net/sfu/servicenow-d2d-j
>> >> > _______________________________________________
>> >> > Kaldi-users mailing list
>> >> > Kal...@li...
>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users
>> >> >
>> >>
>> >>
>> >>
>> >> ------------------------------------------------------------------------------
>> >> How ServiceNow helps IT people transform IT departments:
>> >> 1. A cloud service to automate IT design, transition and operations
>> >> 2. Dashboards that offer high-level views of enterprise services
>> >> 3. A single system of record for all IT processes
>> >> http://p.sf.net/sfu/servicenow-d2d-j
>> >> _______________________________________________
>> >> Kaldi-users mailing list
>> >> Kal...@li...
>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>> >
>> >
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > How ServiceNow helps IT people transform IT departments:
>> > 1. A cloud service to automate IT design, transition and operations
>> > 2. Dashboards that offer high-level views of enterprise services
>> > 3. A single system of record for all IT processes
>> > http://p.sf.net/sfu/servicenow-d2d-j
>> > _______________________________________________
>> > Kaldi-users mailing list
>> > Kal...@li...
>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users
>> >
>>
>>
>> ------------------------------------------------------------------------------
>> How ServiceNow helps IT people transform IT departments:
>> 1. A cloud service to automate IT design, transition and operations
>> 2. Dashboards that offer high-level views of enterprise services
>> 3. A single system of record for all IT processes
>> http://p.sf.net/sfu/servicenow-d2d-j
>> _______________________________________________
>> Kaldi-users mailing list
>> Kal...@li...
>> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
>
>
> ------------------------------------------------------------------------------
> How ServiceNow helps IT people transform IT departments:
> 1. A cloud service to automate IT design, transition and operations
> 2. Dashboards that offer high-level views of enterprise services
> 3. A single system of record for all IT processes
> http://p.sf.net/sfu/servicenow-d2d-j
> _______________________________________________
> Kaldi-users mailing list
> Kal...@li...
> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
|
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 21:49:50
|
My corresponding numbers are -108.535 -> -95.116, and the difference is about the same. [It's strange that the absolute numbers differ slightly, though.] Could you do ls -l data/lang data/lang/phones exp/mono0a exp/mono0a/log exp/mono0a/decode/log and send the output? Dan On Mon, Jun 3, 2013 at 5:42 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Copied from exp/mono0a/log/update.39.log > > LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total > count 849568, avg like/frame -95.5657 > > Hao > > On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > OK, that is quite weird. What is the avg like/frame on the final > > update.??.log? > > Dan > > > > > > > > On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User Communication > and > > Updates <kal...@li...> wrote: > >> > >> Unfortunately those are absolute differences. Here is what I get for the > >> mono0a. > >> > >> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] > >> exp/mono0a/decode_tgpr_eval92/wer_9 > >> > >> And below is the reported one in RESULTS. > >> > >> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, 149 > >> ins, 191 del, 1101 sub ] > >> > >> I will try to checkout the same version as the RESULTS file and run > >> the experiments again. > >> > >> Here is the total data count from exp/mono0a/log/update.0.log. > >> > >> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total > >> count 849568, avg like/frame -109.265 > >> > >> Thanks for your help. > >> > >> Hao > >> > >> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User > >> Communication and Updates <kal...@li...> wrote: > >> > Assuming those are relative numbers, they're probably within the > margin > >> > of > >> > error. See if it's the same on other test sets. Because various > >> > algorithms > >> > call rand(), and different machines implement this differently, > results > >> > aren't fully reproducible. > >> > > >> > But to check that it's not some code change that hurts the results, > you > >> > could try checking out a copy with the same revision number as the > >> > RESULTS > >> > file, and running again and seeing what results you get. If there is > a > >> > difference, I'd like to know. > >> > > >> > It's also possible some files were missing in your WSJ distribution-- > >> > let me > >> > know the data count reported in one the */log/update.log files and > I'll > >> > compare with a local copy. > >> > > >> > Dan > >> > > >> > > >> > > >> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User > Communication > >> > and > >> > Updates <kal...@li...> wrote: > >> >> > >> >> Hi, > >> >> > >> >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce the > >> >> results stated in the RESULTS file. Are we supposed to get exactly > the > >> >> same numbers with the (hyper-)parameters given in the recipe? > >> >> > >> >> Just to give a sense of how far I'm off. My monophone models perform > >> >> 10% > >> >> worse than the reported results on eval92 and the resulting triphone > >> >> models perform 4% worse. > >> >> > >> >> Thanks, > >> >> Hao > >> >> > >> >> > >> >> > >> >> > ------------------------------------------------------------------------------ > >> >> How ServiceNow helps IT people transform IT departments: > >> >> 1. A cloud service to automate IT design, transition and operations > >> >> 2. Dashboards that offer high-level views of enterprise services > >> >> 3. A single system of record for all IT processes > >> >> http://p.sf.net/sfu/servicenow-d2d-j > >> >> _______________________________________________ > >> >> Kaldi-users mailing list > >> >> Kal...@li... > >> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > > >> > > >> > > >> > > ------------------------------------------------------------------------------ > >> > How ServiceNow helps IT people transform IT departments: > >> > 1. A cloud service to automate IT design, transition and operations > >> > 2. Dashboards that offer high-level views of enterprise services > >> > 3. A single system of record for all IT processes > >> > http://p.sf.net/sfu/servicenow-d2d-j > >> > _______________________________________________ > >> > Kaldi-users mailing list > >> > Kal...@li... > >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users > >> > > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 21:42:45
|
Copied from exp/mono0a/log/update.39.log
LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total
count 849568, avg like/frame -95.5657
Hao
On Mon, Jun 3, 2013 at 4:29 PM, Mailing list used for User
Communication and Updates <kal...@li...> wrote:
> OK, that is quite weird. What is the avg like/frame on the final
> update.??.log?
> Dan
>
>
>
> On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User Communication and
> Updates <kal...@li...> wrote:
>>
>> Unfortunately those are absolute differences. Here is what I get for the
>> mono0a.
>>
>> %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ]
>> exp/mono0a/decode_tgpr_eval92/wer_9
>>
>> And below is the reported one in RESULTS.
>>
>> exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, 149
>> ins, 191 del, 1101 sub ]
>>
>> I will try to checkout the same version as the RESULTS file and run
>> the experiments again.
>>
>> Here is the total data count from exp/mono0a/log/update.0.log.
>>
>> LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total
>> count 849568, avg like/frame -109.265
>>
>> Thanks for your help.
>>
>> Hao
>>
>> On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User
>> Communication and Updates <kal...@li...> wrote:
>> > Assuming those are relative numbers, they're probably within the margin
>> > of
>> > error. See if it's the same on other test sets. Because various
>> > algorithms
>> > call rand(), and different machines implement this differently, results
>> > aren't fully reproducible.
>> >
>> > But to check that it's not some code change that hurts the results, you
>> > could try checking out a copy with the same revision number as the
>> > RESULTS
>> > file, and running again and seeing what results you get. If there is a
>> > difference, I'd like to know.
>> >
>> > It's also possible some files were missing in your WSJ distribution--
>> > let me
>> > know the data count reported in one the */log/update.log files and I'll
>> > compare with a local copy.
>> >
>> > Dan
>> >
>> >
>> >
>> > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User Communication
>> > and
>> > Updates <kal...@li...> wrote:
>> >>
>> >> Hi,
>> >>
>> >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce the
>> >> results stated in the RESULTS file. Are we supposed to get exactly the
>> >> same numbers with the (hyper-)parameters given in the recipe?
>> >>
>> >> Just to give a sense of how far I'm off. My monophone models perform
>> >> 10%
>> >> worse than the reported results on eval92 and the resulting triphone
>> >> models perform 4% worse.
>> >>
>> >> Thanks,
>> >> Hao
>> >>
>> >>
>> >>
>> >> ------------------------------------------------------------------------------
>> >> How ServiceNow helps IT people transform IT departments:
>> >> 1. A cloud service to automate IT design, transition and operations
>> >> 2. Dashboards that offer high-level views of enterprise services
>> >> 3. A single system of record for all IT processes
>> >> http://p.sf.net/sfu/servicenow-d2d-j
>> >> _______________________________________________
>> >> Kaldi-users mailing list
>> >> Kal...@li...
>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>> >
>> >
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > How ServiceNow helps IT people transform IT departments:
>> > 1. A cloud service to automate IT design, transition and operations
>> > 2. Dashboards that offer high-level views of enterprise services
>> > 3. A single system of record for all IT processes
>> > http://p.sf.net/sfu/servicenow-d2d-j
>> > _______________________________________________
>> > Kaldi-users mailing list
>> > Kal...@li...
>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users
>> >
>>
>>
>> ------------------------------------------------------------------------------
>> How ServiceNow helps IT people transform IT departments:
>> 1. A cloud service to automate IT design, transition and operations
>> 2. Dashboards that offer high-level views of enterprise services
>> 3. A single system of record for all IT processes
>> http://p.sf.net/sfu/servicenow-d2d-j
>> _______________________________________________
>> Kaldi-users mailing list
>> Kal...@li...
>> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
>
>
> ------------------------------------------------------------------------------
> How ServiceNow helps IT people transform IT departments:
> 1. A cloud service to automate IT design, transition and operations
> 2. Dashboards that offer high-level views of enterprise services
> 3. A single system of record for all IT processes
> http://p.sf.net/sfu/servicenow-d2d-j
> _______________________________________________
> Kaldi-users mailing list
> Kal...@li...
> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
|
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 21:29:51
|
OK, that is quite weird. What is the avg like/frame on the final update.??.log? Dan On Mon, Jun 3, 2013 at 5:00 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Unfortunately those are absolute differences. Here is what I get for the > mono0a. > > %WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ] > exp/mono0a/decode_tgpr_eval92/wer_9 > > And below is the reported one in RESULTS. > > exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, 149 > ins, 191 del, 1101 sub ] > > I will try to checkout the same version as the RESULTS file and run > the experiments again. > > Here is the total data count from exp/mono0a/log/update.0.log. > > LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total > count 849568, avg like/frame -109.265 > > Thanks for your help. > > Hao > > On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User > Communication and Updates <kal...@li...> wrote: > > Assuming those are relative numbers, they're probably within the margin > of > > error. See if it's the same on other test sets. Because various > algorithms > > call rand(), and different machines implement this differently, results > > aren't fully reproducible. > > > > But to check that it's not some code change that hurts the results, you > > could try checking out a copy with the same revision number as the > RESULTS > > file, and running again and seeing what results you get. If there is a > > difference, I'd like to know. > > > > It's also possible some files were missing in your WSJ distribution-- > let me > > know the data count reported in one the */log/update.log files and I'll > > compare with a local copy. > > > > Dan > > > > > > > > On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User Communication > and > > Updates <kal...@li...> wrote: > >> > >> Hi, > >> > >> I'm following the WSJ s5 recipe, and I wasn't able to reproduce the > >> results stated in the RESULTS file. Are we supposed to get exactly the > >> same numbers with the (hyper-)parameters given in the recipe? > >> > >> Just to give a sense of how far I'm off. My monophone models perform 10% > >> worse than the reported results on eval92 and the resulting triphone > >> models perform 4% worse. > >> > >> Thanks, > >> Hao > >> > >> > >> > ------------------------------------------------------------------------------ > >> How ServiceNow helps IT people transform IT departments: > >> 1. A cloud service to automate IT design, transition and operations > >> 2. Dashboards that offer high-level views of enterprise services > >> 3. A single system of record for all IT processes > >> http://p.sf.net/sfu/servicenow-d2d-j > >> _______________________________________________ > >> Kaldi-users mailing list > >> Kal...@li... > >> https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > > > > > ------------------------------------------------------------------------------ > > How ServiceNow helps IT people transform IT departments: > > 1. A cloud service to automate IT design, transition and operations > > 2. Dashboards that offer high-level views of enterprise services > > 3. A single system of record for all IT processes > > http://p.sf.net/sfu/servicenow-d2d-j > > _______________________________________________ > > Kaldi-users mailing list > > Kal...@li... > > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 21:00:27
|
Unfortunately those are absolute differences. Here is what I get for the mono0a.
%WER 40.44 [ 2282 / 5643, 118 ins, 537 del, 1627 sub ]
exp/mono0a/decode_tgpr_eval92/wer_9
And below is the reported one in RESULTS.
exp/mono0a/decode_tgpr_eval92/wer_10:%WER 25.54 [ 1441 / 5643, 149
ins, 191 del, 1101 sub ]
I will try to checkout the same version as the RESULTS file and run
the experiments again.
Here is the total data count from exp/mono0a/log/update.0.log.
LOG (gmm-sum-accs:main():gmm-sum-accs.cc:60) Summed 2 stats, total
count 849568, avg like/frame -109.265
Thanks for your help.
Hao
On Mon, Jun 3, 2013 at 3:43 PM, Mailing list used for User
Communication and Updates <kal...@li...> wrote:
> Assuming those are relative numbers, they're probably within the margin of
> error. See if it's the same on other test sets. Because various algorithms
> call rand(), and different machines implement this differently, results
> aren't fully reproducible.
>
> But to check that it's not some code change that hurts the results, you
> could try checking out a copy with the same revision number as the RESULTS
> file, and running again and seeing what results you get. If there is a
> difference, I'd like to know.
>
> It's also possible some files were missing in your WSJ distribution-- let me
> know the data count reported in one the */log/update.log files and I'll
> compare with a local copy.
>
> Dan
>
>
>
> On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User Communication and
> Updates <kal...@li...> wrote:
>>
>> Hi,
>>
>> I'm following the WSJ s5 recipe, and I wasn't able to reproduce the
>> results stated in the RESULTS file. Are we supposed to get exactly the
>> same numbers with the (hyper-)parameters given in the recipe?
>>
>> Just to give a sense of how far I'm off. My monophone models perform 10%
>> worse than the reported results on eval92 and the resulting triphone
>> models perform 4% worse.
>>
>> Thanks,
>> Hao
>>
>>
>> ------------------------------------------------------------------------------
>> How ServiceNow helps IT people transform IT departments:
>> 1. A cloud service to automate IT design, transition and operations
>> 2. Dashboards that offer high-level views of enterprise services
>> 3. A single system of record for all IT processes
>> http://p.sf.net/sfu/servicenow-d2d-j
>> _______________________________________________
>> Kaldi-users mailing list
>> Kal...@li...
>> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
>
>
> ------------------------------------------------------------------------------
> How ServiceNow helps IT people transform IT departments:
> 1. A cloud service to automate IT design, transition and operations
> 2. Dashboards that offer high-level views of enterprise services
> 3. A single system of record for all IT processes
> http://p.sf.net/sfu/servicenow-d2d-j
> _______________________________________________
> Kaldi-users mailing list
> Kal...@li...
> https://lists.sourceforge.net/lists/listinfo/kaldi-users
>
|
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 20:43:19
|
Assuming those are relative numbers, they're probably within the margin of error. See if it's the same on other test sets. Because various algorithms call rand(), and different machines implement this differently, results aren't fully reproducible. But to check that it's not some code change that hurts the results, you could try checking out a copy with the same revision number as the RESULTS file, and running again and seeing what results you get. If there is a difference, I'd like to know. It's also possible some files were missing in your WSJ distribution-- let me know the data count reported in one the */log/update.log files and I'll compare with a local copy. Dan On Mon, Jun 3, 2013 at 4:34 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Hi, > > I'm following the WSJ s5 recipe, and I wasn't able to reproduce the > results stated in the RESULTS file. Are we supposed to get exactly the > same numbers with the (hyper-)parameters given in the recipe? > > Just to give a sense of how far I'm off. My monophone models perform 10% > worse than the reported results on eval92 and the resulting triphone > models perform 4% worse. > > Thanks, > Hao > > > ------------------------------------------------------------------------------ > How ServiceNow helps IT people transform IT departments: > 1. A cloud service to automate IT design, transition and operations > 2. Dashboards that offer high-level views of enterprise services > 3. A single system of record for all IT processes > http://p.sf.net/sfu/servicenow-d2d-j > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-06-03 20:34:09
|
Hi, I'm following the WSJ s5 recipe, and I wasn't able to reproduce the results stated in the RESULTS file. Are we supposed to get exactly the same numbers with the (hyper-)parameters given in the recipe? Just to give a sense of how far I'm off. My monophone models perform 10% worse than the reported results on eval92 and the resulting triphone models perform 4% worse. Thanks, Hao |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-05-25 06:14:56
|
Hi, please understand that the convert_htk.sh is not guaranteed to work for all HTK models. You should build everything from scratch in Kaldi. Also, the decision tree format is not designed for human readability. -Arnab On Sat, May 25, 2013 at 6:59 AM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > Hi, > i am trying to use convert_htk.sh to convert htk acoustic model to > kaldi,because my acoustic model is a little complicated. > i had some problem in producing kaldi.tree. > Can someone mail me your kaldi.tree please? > i will under stand the format of kaldi.tree. > > ------------------------------------------------------------------------------ > Try New Relic Now & We'll Send You this Cool Shirt > New Relic is the only SaaS-based application performance monitoring service > that delivers powerful full stack analytics. Optimize and monitor your > browser, app, & servers with just a few lines of code. Try New Relic > and get this awesome Nerd Life shirt! http://p.sf.net/sfu/newrelic_d2d_may > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-05-25 05:59:34
|
Hi, i am trying to use convert_htk.sh to convert htk acoustic model to kaldi,because my acoustic model is a little complicated. i had some problem in producing kaldi.tree. Can someone mail me your kaldi.tree please? i will under stand the format of kaldi.tree. |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-05-14 16:30:16
|
I just committed the changes. Dan On Tue, May 14, 2013 at 12:29 PM, Mailing list used for User Communication and Updates <kal...@li...> wrote: > When I will finish the Python wrapper and writing the docs, I will publish > it probably on www.bitbucket.org like a git repository. > I can drop a link somewhere if anybody would be interested. > > Ondra > > > On 14 May 2013 18:26, Ondrej Platek <ond...@gm...> wrote: > >> Good to hear. I am trying from last week to learn how was it designed. >> When are you going to commit the changes? In days in weeks? >> I would like to use the new code to build upon. >> >> I would like to run some experiments first, >> than I need to build a python wrapper (my project requirements). >> Currently I am finishing a python wrapper for the compute-mfcc-featscc , >> gmm-latgen-faster.cc and lattice-best-path.cc >> using https://cffi.readthedocs.org/en/release-0.6/. >> >> Basically I am trying to build shared library containing >> functionality of the compute-mfcc-featscc , gmm-latgen-faster.cc and >> lattice-best-path.cc. >> Than the rest is very easy using python-cffi ( >> https://cffi.readthedocs.org/en/release-0.6/). >> I guess there is no plan to distribute Kaldi as shared library, right? >> (Pity for me) >> >> Ondra >> >> >> On 14 May 2013 18:13, Mailing list used for User Communication and >> Updates <kal...@li...> wrote: >> >>> By the way, since we are talking about the online decoder, I should >>> mention that I am just about to commit some changes, involving a >>> reorganization of the internal code and the addition of unit-tests. Vassil >>> has checked that it still works for his demo. >>> >>> Dan >>> >>> >>> >>> On Tue, May 14, 2013 at 12:09 PM, Daniel Povey <dp...@gm...> wrote: >>> >>>> You have to be careful that you give the online decoder the right >>>> options corresponding tot the features that you used. See if there are any >>>> non-default options in your MFCC extraction or PLP extraction, for >>>> instance, and that the LDA window is set correctly. >>>> >>>> Dan >>>> >>>> >>>> >>>> On Tue, May 14, 2013 at 10:10 AM, Mailing list used for User >>>> Communication and Updates <kal...@li...> wrote: >>>> >>>>> As I said on my data have results like 70% WER. It is due to some bug, >>>>> which I have not fixed yet. >>>>> >>>>> In the online demo I tried to use our test (quite general data - in >>>>> terms of LM) and test it against online demo, >>>>> because as far as I understood it the online demo was trained on the >>>>> same data which are in supplied as the test data. >>>>> >>>>> Basically, I was asking for the numbers because I need to know if I >>>>> have still bug in my script or it is just normal performance. >>>>> >>>>> Thansk >>>>> >>>>> Ondra >>>>> >>>>> >>>>> >>>>> >>>>> On 14 May 2013 15:39, Mailing list used for User Communication and >>>>> Updates <kal...@li...> wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> I don't have specific numbers to share (as I said the online decoder >>>>>> is not rigorously tested as far as I know). In principle I think the >>>>>> online decoder should perform slightly worse than the offline versions >>>>>> due to things like online CMN computation. >>>>>> >>>>>> What do you mean exactly when you say that the results for the online >>>>>> demo are not so good? As far as I remember the WER for the test WAV >>>>>> files was about 7%, which is not so bad as today's speech recognition >>>>>> goes. Of course the WER in this case is so low, because the language >>>>>> model is relatively small and perfectly matches the test files (the LM >>>>>> is trained on the text corresponding to the audio book from which the >>>>>> test utterances were taken - not the best practice of course). >>>>>> >>>>>> Vassil >>>>>> >>>>>> On Tue, May 14, 2013 at 2:59 PM, Mailing list used for User >>>>>> Communication and Updates <kal...@li...> wrote: >>>>>> > Hi Vassil, >>>>>> > >>>>>> > May I asked about the performance (in terms of WER/SER) of online >>>>>> decoder? >>>>>> > >>>>>> > What is the expected WER for online-wav-gmm-decode-faster, >>>>>> > if I reached with gmm-latgen-faster scores in table below? >>>>>> > I would like to use the same trained models for >>>>>> online-wav-gmm-decode-faster >>>>>> > as >>>>>> > I used for gmm-latgen-faster (and lattice-best-path). >>>>>> > >>>>>> > exp RT coef WER >>>>>> > _ri3b_fmmi_b 11.33336625 19.42 >>>>>> > tri2b_mpe 1.2307175 21.22 >>>>>> > mono 2.8952 49.88 >>>>>> > tri3b_mmi 1.24198625 20.01 >>>>>> > tri1 2.398165 28.13 >>>>>> > tri2b_mmi 1.65662125 20.34 >>>>>> > tri2a 2.15838 28.43 >>>>>> > tri2b 1.451105 27.96 >>>>>> > tri3b 1.0596075 27.79 >>>>>> > tri3b_fmmi_c 11.4361875 19.13 >>>>>> > tri2b_mmi_b0.05 2.0089575 19.64 >>>>>> > tri3b_fmmi_d 16.1227625 19.5 >>>>>> > >>>>>> > The names and the core settings of experiments are the same as in >>>>>> > egs/voxforge/s5/run.sh >>>>>> > >>>>>> > >>>>>> > With the online-wav-gmm-decode-faster >>>>>> > I reached really bad scores due to bugs (which I still need to fix), >>>>>> > but still also in the online demo the results are not so nice. >>>>>> > >>>>>> > Thanks for any help >>>>>> > >>>>>> > Ondra Platek >>>>>> > >>>>>> > >>>>>> > On 26 April 2013 15:35, Mailing list used for User Communication >>>>>> and Updates >>>>>> > <kal...@li...> wrote: >>>>>> >> >>>>>> >> Hi, >>>>>> >> as for taking input from a microphone you might want to have a >>>>>> look at >>>>>> >> the (optional) online decoder. There is a simple demo script in >>>>>> >> egs/voxforge/online-demo. There is also a version that can decode >>>>>> .wav >>>>>> >> files without a separate feature extraction step. The code is not >>>>>> >> extensively tested and lacks some polish and flexibility, but >>>>>> perhaps >>>>>> >> you can use it as an example. >>>>>> >> >>>>>> >> Vassil >>>>>> >> >>>>>> >> On Fri, Apr 26, 2013 at 3:55 PM, Mailing list used for User >>>>>> >> Communication and Updates <kal...@li...> >>>>>> wrote: >>>>>> >> > Dear all, >>>>>> >> > I'm working on the development of an R&D voice-to-voice module >>>>>> and I >>>>>> >> > would >>>>>> >> > like to use Kaldi for the transcription part (we made some tests >>>>>> with >>>>>> >> > PocketSphinx but we found that Kaldi is more adapted for our >>>>>> task). >>>>>> >> > >>>>>> >> > The problem is that I'm not really expert in speech2text (Kaldi >>>>>> models >>>>>> >> > will >>>>>> >> > be provided by another team) and I did not found scripts (or >>>>>> plugins) >>>>>> >> > allowing an easy integration of Kaldi : How to manage streams >>>>>> (with >>>>>> >> > gstreammer for example)? how we can decode a single wav file? >>>>>> etc. >>>>>> >> > >>>>>> >> > Is someone else has already handle these problems? are there any >>>>>> scripts >>>>>> >> > to >>>>>> >> > facilitate this task? >>>>>> >> > >>>>>> >> > thanks in advance for your reply, >>>>>> >> > >>>>>> >> > Best regards >>>>>> >> > >>>>>> >> > >>>>>> >> > >>>>>> >> > >>>>>> ------------------------------------------------------------------------------ >>>>>> >> > Try New Relic Now & We'll Send You this Cool Shirt >>>>>> >> > New Relic is the only SaaS-based application performance >>>>>> monitoring >>>>>> >> > service >>>>>> >> > that delivers powerful full stack analytics. Optimize and >>>>>> monitor your >>>>>> >> > browser, app, & servers with just a few lines of code. Try New >>>>>> Relic >>>>>> >> > and get this awesome Nerd Life shirt! >>>>>> >> > http://p.sf.net/sfu/newrelic_d2d_apr >>>>>> >> > _______________________________________________ >>>>>> >> > Kaldi-users mailing list >>>>>> >> > Kal...@li... >>>>>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>>> >> > >>>>>> >> >>>>>> >> >>>>>> >> >>>>>> ------------------------------------------------------------------------------ >>>>>> >> Try New Relic Now & We'll Send You this Cool Shirt >>>>>> >> New Relic is the only SaaS-based application performance monitoring >>>>>> >> service >>>>>> >> that delivers powerful full stack analytics. Optimize and monitor >>>>>> your >>>>>> >> browser, app, & servers with just a few lines of code. Try New >>>>>> Relic >>>>>> >> and get this awesome Nerd Life shirt! >>>>>> http://p.sf.net/sfu/newrelic_d2d_apr >>>>>> >> _______________________________________________ >>>>>> >> Kaldi-users mailing list >>>>>> >> Kal...@li... >>>>>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>>> > >>>>>> > >>>>>> > >>>>>> > >>>>>> ------------------------------------------------------------------------------ >>>>>> > AlienVault Unified Security Management (USM) platform delivers >>>>>> complete >>>>>> > security visibility with the essential security capabilities. >>>>>> Easily and >>>>>> > efficiently configure, manage, and operate all of your security >>>>>> controls >>>>>> > from a single console and one unified framework. Download a free >>>>>> trial. >>>>>> > http://p.sf.net/sfu/alienvault_d2d >>>>>> > _______________________________________________ >>>>>> > Kaldi-users mailing list >>>>>> > Kal...@li... >>>>>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>>> > >>>>>> >>>>>> >>>>>> ------------------------------------------------------------------------------ >>>>>> AlienVault Unified Security Management (USM) platform delivers >>>>>> complete >>>>>> security visibility with the essential security capabilities. Easily >>>>>> and >>>>>> efficiently configure, manage, and operate all of your security >>>>>> controls >>>>>> from a single console and one unified framework. Download a free >>>>>> trial. >>>>>> http://p.sf.net/sfu/alienvault_d2d >>>>>> _______________________________________________ >>>>>> Kaldi-users mailing list >>>>>> Kal...@li... >>>>>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>>> >>>>> >>>>> >>>>> >>>>> ------------------------------------------------------------------------------ >>>>> AlienVault Unified Security Management (USM) platform delivers complete >>>>> security visibility with the essential security capabilities. Easily >>>>> and >>>>> efficiently configure, manage, and operate all of your security >>>>> controls >>>>> from a single console and one unified framework. Download a free trial. >>>>> http://p.sf.net/sfu/alienvault_d2d >>>>> _______________________________________________ >>>>> Kaldi-users mailing list >>>>> Kal...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>> >>>>> >>>> >>> >>> >>> ------------------------------------------------------------------------------ >>> AlienVault Unified Security Management (USM) platform delivers complete >>> security visibility with the essential security capabilities. Easily and >>> efficiently configure, manage, and operate all of your security controls >>> from a single console and one unified framework. Download a free trial. >>> http://p.sf.net/sfu/alienvault_d2d >>> _______________________________________________ >>> Kaldi-users mailing list >>> Kal...@li... >>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>> >>> >> > > > ------------------------------------------------------------------------------ > AlienVault Unified Security Management (USM) platform delivers complete > security visibility with the essential security capabilities. Easily and > efficiently configure, manage, and operate all of your security controls > from a single console and one unified framework. Download a free trial. > http://p.sf.net/sfu/alienvault_d2d > _______________________________________________ > Kaldi-users mailing list > Kal...@li... > https://lists.sourceforge.net/lists/listinfo/kaldi-users > > |
|
From: Mailing l. u. f. U. C. a. U. <kal...@li...> - 2013-05-14 16:29:23
|
When I will finish the Python wrapper and writing the docs, I will publish it probably on www.bitbucket.org like a git repository. I can drop a link somewhere if anybody would be interested. Ondra On 14 May 2013 18:26, Ondrej Platek <ond...@gm...> wrote: > Good to hear. I am trying from last week to learn how was it designed. > When are you going to commit the changes? In days in weeks? > I would like to use the new code to build upon. > > I would like to run some experiments first, > than I need to build a python wrapper (my project requirements). > Currently I am finishing a python wrapper for the compute-mfcc-featscc , > gmm-latgen-faster.cc and lattice-best-path.cc > using https://cffi.readthedocs.org/en/release-0.6/. > > Basically I am trying to build shared library containing > functionality of the compute-mfcc-featscc , gmm-latgen-faster.cc and > lattice-best-path.cc. > Than the rest is very easy using python-cffi ( > https://cffi.readthedocs.org/en/release-0.6/). > I guess there is no plan to distribute Kaldi as shared library, right? > (Pity for me) > > Ondra > > > On 14 May 2013 18:13, Mailing list used for User Communication and Updates > <kal...@li...> wrote: > >> By the way, since we are talking about the online decoder, I should >> mention that I am just about to commit some changes, involving a >> reorganization of the internal code and the addition of unit-tests. Vassil >> has checked that it still works for his demo. >> >> Dan >> >> >> >> On Tue, May 14, 2013 at 12:09 PM, Daniel Povey <dp...@gm...> wrote: >> >>> You have to be careful that you give the online decoder the right >>> options corresponding tot the features that you used. See if there are any >>> non-default options in your MFCC extraction or PLP extraction, for >>> instance, and that the LDA window is set correctly. >>> >>> Dan >>> >>> >>> >>> On Tue, May 14, 2013 at 10:10 AM, Mailing list used for User >>> Communication and Updates <kal...@li...> wrote: >>> >>>> As I said on my data have results like 70% WER. It is due to some bug, >>>> which I have not fixed yet. >>>> >>>> In the online demo I tried to use our test (quite general data - in >>>> terms of LM) and test it against online demo, >>>> because as far as I understood it the online demo was trained on the >>>> same data which are in supplied as the test data. >>>> >>>> Basically, I was asking for the numbers because I need to know if I >>>> have still bug in my script or it is just normal performance. >>>> >>>> Thansk >>>> >>>> Ondra >>>> >>>> >>>> >>>> >>>> On 14 May 2013 15:39, Mailing list used for User Communication and >>>> Updates <kal...@li...> wrote: >>>> >>>>> Hi, >>>>> >>>>> I don't have specific numbers to share (as I said the online decoder >>>>> is not rigorously tested as far as I know). In principle I think the >>>>> online decoder should perform slightly worse than the offline versions >>>>> due to things like online CMN computation. >>>>> >>>>> What do you mean exactly when you say that the results for the online >>>>> demo are not so good? As far as I remember the WER for the test WAV >>>>> files was about 7%, which is not so bad as today's speech recognition >>>>> goes. Of course the WER in this case is so low, because the language >>>>> model is relatively small and perfectly matches the test files (the LM >>>>> is trained on the text corresponding to the audio book from which the >>>>> test utterances were taken - not the best practice of course). >>>>> >>>>> Vassil >>>>> >>>>> On Tue, May 14, 2013 at 2:59 PM, Mailing list used for User >>>>> Communication and Updates <kal...@li...> wrote: >>>>> > Hi Vassil, >>>>> > >>>>> > May I asked about the performance (in terms of WER/SER) of online >>>>> decoder? >>>>> > >>>>> > What is the expected WER for online-wav-gmm-decode-faster, >>>>> > if I reached with gmm-latgen-faster scores in table below? >>>>> > I would like to use the same trained models for >>>>> online-wav-gmm-decode-faster >>>>> > as >>>>> > I used for gmm-latgen-faster (and lattice-best-path). >>>>> > >>>>> > exp RT coef WER >>>>> > _ri3b_fmmi_b 11.33336625 19.42 >>>>> > tri2b_mpe 1.2307175 21.22 >>>>> > mono 2.8952 49.88 >>>>> > tri3b_mmi 1.24198625 20.01 >>>>> > tri1 2.398165 28.13 >>>>> > tri2b_mmi 1.65662125 20.34 >>>>> > tri2a 2.15838 28.43 >>>>> > tri2b 1.451105 27.96 >>>>> > tri3b 1.0596075 27.79 >>>>> > tri3b_fmmi_c 11.4361875 19.13 >>>>> > tri2b_mmi_b0.05 2.0089575 19.64 >>>>> > tri3b_fmmi_d 16.1227625 19.5 >>>>> > >>>>> > The names and the core settings of experiments are the same as in >>>>> > egs/voxforge/s5/run.sh >>>>> > >>>>> > >>>>> > With the online-wav-gmm-decode-faster >>>>> > I reached really bad scores due to bugs (which I still need to fix), >>>>> > but still also in the online demo the results are not so nice. >>>>> > >>>>> > Thanks for any help >>>>> > >>>>> > Ondra Platek >>>>> > >>>>> > >>>>> > On 26 April 2013 15:35, Mailing list used for User Communication and >>>>> Updates >>>>> > <kal...@li...> wrote: >>>>> >> >>>>> >> Hi, >>>>> >> as for taking input from a microphone you might want to have a look >>>>> at >>>>> >> the (optional) online decoder. There is a simple demo script in >>>>> >> egs/voxforge/online-demo. There is also a version that can decode >>>>> .wav >>>>> >> files without a separate feature extraction step. The code is not >>>>> >> extensively tested and lacks some polish and flexibility, but >>>>> perhaps >>>>> >> you can use it as an example. >>>>> >> >>>>> >> Vassil >>>>> >> >>>>> >> On Fri, Apr 26, 2013 at 3:55 PM, Mailing list used for User >>>>> >> Communication and Updates <kal...@li...> >>>>> wrote: >>>>> >> > Dear all, >>>>> >> > I'm working on the development of an R&D voice-to-voice module >>>>> and I >>>>> >> > would >>>>> >> > like to use Kaldi for the transcription part (we made some tests >>>>> with >>>>> >> > PocketSphinx but we found that Kaldi is more adapted for our >>>>> task). >>>>> >> > >>>>> >> > The problem is that I'm not really expert in speech2text (Kaldi >>>>> models >>>>> >> > will >>>>> >> > be provided by another team) and I did not found scripts (or >>>>> plugins) >>>>> >> > allowing an easy integration of Kaldi : How to manage streams >>>>> (with >>>>> >> > gstreammer for example)? how we can decode a single wav file? etc. >>>>> >> > >>>>> >> > Is someone else has already handle these problems? are there any >>>>> scripts >>>>> >> > to >>>>> >> > facilitate this task? >>>>> >> > >>>>> >> > thanks in advance for your reply, >>>>> >> > >>>>> >> > Best regards >>>>> >> > >>>>> >> > >>>>> >> > >>>>> >> > >>>>> ------------------------------------------------------------------------------ >>>>> >> > Try New Relic Now & We'll Send You this Cool Shirt >>>>> >> > New Relic is the only SaaS-based application performance >>>>> monitoring >>>>> >> > service >>>>> >> > that delivers powerful full stack analytics. Optimize and monitor >>>>> your >>>>> >> > browser, app, & servers with just a few lines of code. Try New >>>>> Relic >>>>> >> > and get this awesome Nerd Life shirt! >>>>> >> > http://p.sf.net/sfu/newrelic_d2d_apr >>>>> >> > _______________________________________________ >>>>> >> > Kaldi-users mailing list >>>>> >> > Kal...@li... >>>>> >> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>> >> > >>>>> >> >>>>> >> >>>>> >> >>>>> ------------------------------------------------------------------------------ >>>>> >> Try New Relic Now & We'll Send You this Cool Shirt >>>>> >> New Relic is the only SaaS-based application performance monitoring >>>>> >> service >>>>> >> that delivers powerful full stack analytics. Optimize and monitor >>>>> your >>>>> >> browser, app, & servers with just a few lines of code. Try New Relic >>>>> >> and get this awesome Nerd Life shirt! >>>>> http://p.sf.net/sfu/newrelic_d2d_apr >>>>> >> _______________________________________________ >>>>> >> Kaldi-users mailing list >>>>> >> Kal...@li... >>>>> >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>> > >>>>> > >>>>> > >>>>> > >>>>> ------------------------------------------------------------------------------ >>>>> > AlienVault Unified Security Management (USM) platform delivers >>>>> complete >>>>> > security visibility with the essential security capabilities. Easily >>>>> and >>>>> > efficiently configure, manage, and operate all of your security >>>>> controls >>>>> > from a single console and one unified framework. Download a free >>>>> trial. >>>>> > http://p.sf.net/sfu/alienvault_d2d >>>>> > _______________________________________________ >>>>> > Kaldi-users mailing list >>>>> > Kal...@li... >>>>> > https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>> > >>>>> >>>>> >>>>> ------------------------------------------------------------------------------ >>>>> AlienVault Unified Security Management (USM) platform delivers complete >>>>> security visibility with the essential security capabilities. Easily >>>>> and >>>>> efficiently configure, manage, and operate all of your security >>>>> controls >>>>> from a single console and one unified framework. Download a free trial. >>>>> http://p.sf.net/sfu/alienvault_d2d >>>>> _______________________________________________ >>>>> Kaldi-users mailing list >>>>> Kal...@li... >>>>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>>> >>>> >>>> >>>> >>>> ------------------------------------------------------------------------------ >>>> AlienVault Unified Security Management (USM) platform delivers complete >>>> security visibility with the essential security capabilities. Easily and >>>> efficiently configure, manage, and operate all of your security controls >>>> from a single console and one unified framework. Download a free trial. >>>> http://p.sf.net/sfu/alienvault_d2d >>>> _______________________________________________ >>>> Kaldi-users mailing list >>>> Kal...@li... >>>> https://lists.sourceforge.net/lists/listinfo/kaldi-users >>>> >>>> >>> >> >> >> ------------------------------------------------------------------------------ >> AlienVault Unified Security Management (USM) platform delivers complete >> security visibility with the essential security capabilities. Easily and >> efficiently configure, manage, and operate all of your security controls >> from a single console and one unified framework. Download a free trial. >> http://p.sf.net/sfu/alienvault_d2d >> _______________________________________________ >> Kaldi-users mailing list >> Kal...@li... >> https://lists.sourceforge.net/lists/listinfo/kaldi-users >> >> > |
1 message has been excluded from this view by a project administrator.
