Menu

Baum Welch stops in first iteration

Help
meshweta_7
2013-04-25
2013-04-25
  • meshweta_7

    meshweta_7 - 2013-04-25

    I am doing isolated word recognition and while executing RunAll.pl it stops in the first iteration itself.
    log file details:

    INFO: main.c(196): Compiled on Mar 31 2013 at 14:39:33
    /unused/asrws/workspace/hmm/bin/bw \ -moddeffn /unused/asrws/workspace/hmm/model_architecture/t.ci.mdef \ -ts2cbfn .cont. \ -mixwfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/mixture_weights \ -mwfloor 1e-08 \ -tmatfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/transition_matrices \ -meanfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/means \ -varfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/variances \ -ltsoov no \ -dictfn /unused/asrws/workspace/hmm/etc/t.dic \ -fdictfn /unused/asrws/workspace/hmm/etc/t.filler \ -ctlfn /unused/asrws/workspace/hmm/etc/t_train.fileids \ -part 1 \ -npart 1 \ -cepdir /unused/asrws/workspace/hmm/feat \ -cepext mfc \ -lsnfn /unused/asrws/workspace/hmm/etc/t_train.transcription \ -accumdir /unused/asrws/workspace/hmm/bwaccumdir/t_buff_1 \ -varfloor 0.0001 \ -topn 1 \ -abeam 1e-90 \ -bbeam 1e-10 \ -agc none \ -cmn current \ -varnorm no \ -meanreest yes \ -varreest yes -2passvar no \ -tmatreest yes \ -fullvar no \ -diagfull no \ -feat 1s_c_d_dd \ -ceplen 13 \ -timing no

    [Switch] [Default] [Value]
    -help no no
    -example no no
    -hmmdir
    -moddeffn /unused/asrws/workspace/hmm/model_architecture/t.ci.mdef
    -tmatfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/transition_matrices
    -mixwfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/mixture_weights
    -meanfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/means
    -varfn /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/variances
    -fullvar no no
    -diagfull no no
    -mwfloor 0.00001 1.000000e-08
    -tpfloor 0.0001 1.000000e-04
    -varfloor 0.00001 1.000000e-04
    -topn 4 1
    -dictfn /unused/asrws/workspace/hmm/etc/t.dic
    -fdictfn /unused/asrws/workspace/hmm/etc/t.filler
    -ltsoov no no
    -ctlfn /unused/asrws/workspace/hmm/etc/t_train.fileids
    -nskip
    -runlen -1 -1
    -part 1
    -npart 1
    -cepext mfc mfc
    -cepdir /unused/asrws/workspace/hmm/feat
    -phsegext phseg phseg
    -phsegdir
    -outphsegdir
    -sentdir
    -sentext sent sent
    -lsnfn /unused/asrws/workspace/hmm/etc/t_train.transcription
    -accumdir /unused/asrws/workspace/hmm/bwaccumdir/t_buff_1
    -ceplen 13 13
    -cepwin 0 0
    -agc max none
    -cmn current current
    -varnorm no no
    -silcomp none none
    -sildel no no
    -siltag SIL SIL
    -abeam 1e-100 1.000000e-90
    -bbeam 1e-100 1.000000e-10
    -varreest yes yes
    -meanreest yes yes
    -mixwreest yes yes
    -tmatreest yes yes
    -mllrmat
    -cb2mllrfn .1cls. .1cls.
    -ts2cbfn .cont.
    -feat 1s_c_d_dd 1s_c_d_dd
    -svspec
    -ldafn
    -ldadim 29 29
    -ldaaccum no no
    -timing yes no
    -viterbi no no
    -2passvar no no
    -sildelfn
    -spthresh 0.0 0.000000e+00
    -maxuttlen 0 0
    -ckptintv
    -outputfullpath no no
    -fullsuffixmatch no no
    -pdumpdir
    INFO: main.c(255): Reading /unused/asrws/workspace/hmm/model_architecture/t.ci.mdef
    INFO: model_def_io.c(587): Model definition info:
    INFO: model_def_io.c(588): 63 total models defined (63 base, 0 tri)
    INFO: model_def_io.c(589): 252 total states
    INFO: model_def_io.c(590): 189 total tied states
    INFO: model_def_io.c(591): 189 total tied CI states
    INFO: model_def_io.c(592): 63 total tied transition matrices
    INFO: model_def_io.c(593): 4 max state/model
    INFO: model_def_io.c(594): 4 min state/model
    INFO: s3mixw_io.c(116): Read /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/mixture_weights [189x1x1 array]
    INFO: s3tmat_io.c(115): Read /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/transition_matrices [63x3x4 array]
    INFO: mod_inv.c(297): inserting tprob floor 1.000000e-04 and renormalizing
    INFO: s3gau_io.c(166): Read /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/means [189x1x1 array]
    INFO: s3gau_io.c(166): Read /unused/asrws/workspace/hmm/model_parameters/t.ci_cont_flatinitial/variances [189x1x1 array]
    INFO: gauden.c(181): 189 total mgau
    INFO: gauden.c(155): 1 feature streams (|0|=39 )
    INFO: gauden.c(192): 1 total densities
    INFO: gauden.c(98): min_var=1.000000e-04
    INFO: gauden.c(170): compute 1 densities/frame
    INFO: main.c(363): Will reestimate mixing weights.
    INFO: main.c(365): Will reestimate means.
    INFO: main.c(367): Will reestimate variances.
    INFO: main.c(369): WIll NOT optionally delete silence in Baum Welch or Viterbi.
    INFO: main.c(377): Will reestimate transition matrices
    INFO: main.c(390): Reading main lexicon: /unused/asrws/workspace/hmm/etc/t.dic
    INFO: lexicon.c(233): 135 entries added from /unused/asrws/workspace/hmm/etc/t.dic
    INFO: main.c(402): Reading filler lexicon: /unused/asrws/workspace/hmm/etc/t.filler
    INFO: lexicon.c(233): 3 entries added from /unused/asrws/workspace/hmm/etc/t.filler
    INFO: main.c(423): Silence Tag SIL
    INFO: corpus.c(1343): Will process all remaining utts starting at 0
    INFO: main.c(622): Reestimation: Baum-Welch
    column defns
    <seq>
    <id>
    <n_frame_in>
    <n_frame_del>
    <n_state_shmm>
    <avg_states_alpha>
    <avg_states_beta>
    <avg_states_reest>
    <avg_posterior_prune>
    <frame_log_lik>
    <utt_log_lik>
    ... timing info ...
    utt> 0 aasaam1 78 0WARNING: "corpus.c", line 1986: LSN utt id, amdhrpradesh1, does not match ctl utt id, aasaam1.
    INFO: cvt2triphone.c(199): no multiphones defined, no conversion done
    48 37 17 34 1.683315e-11 -6.891933e+00 -5.375708e+02
    utt> 1 amdhrpradesh1 105 0WARNING: "corpus.c", line 1986: LSN utt id, aasaam1, does not match ctl utt id, amdhrpradesh1.
    24 21 14 27 1.102779e-11 -9.639659e+00 -1.012164e+03
    utt> 2 adxqaakuu1 87 0 28 24 15 29 1.268100e-11 -1.157078e+01 -1.006658e+03
    utt> 3 alaamqg1 94 0 28 24 15 29 1.001631e-11 -6.376739e+00 -5.994135e+02
    utt> 4 amdxmaannikobaar1 127 0 60 49 23 46 2.812069e-11 -8.461900e+00 -1.074661e+03
    utt> 5 amaranaath1 77 0 36 29 16 32 1.263283e-11 -6.660715e+00 -5.128751e+02
    utt> 6 ambaajii1 75 0 32 26 15 30 1.533491e-11 -7.877924e+00 -5.908443e+02
    utt> 7 arunachalpradesh1 117 0 60 48 22 44 2.430301e-11 -1.321694e+01 -1.546382e+03
    utt> 8 bastar1 62 0 32 25 14 27 8.422089e-12 -1.402400e+01 -8.694883e+02
    utt> 9 bhadraacalam1 125 0 40 35 19 38 1.835160e-11 -1.203149e+01 -1.503936e+03
    utt> 10 bihaar1 73 0 28 24 14 27 1.062759e-11 -9.571336e+00 -6.987075e+02
    utt> 11 bodhagayaa1 73 0 40 31 16 32 1.393580e-11 -9.796325e+00 -7.151317e+02
    utt> 12 caaramiinaar1 89 0 40 33 17 35 1.599323e-11 -9.119014e+00 -8.115922e+02
    utt> 13 camdxiigadxhq1 113 0 40 34 19 37 2.289047e-11 -8.763392e+00 -9.902633e+02
    utt> 14 chamba1 60 0 28 23 13 26 9.390759e-12 -1.306429e+01 -7.838576e+02
    utt> 15 caraareshariiphql1 111 0 48 40 20 40 2.387723e-11 -7.371505e+00 -8.182370e+02
    utt> 16 chashmeshaahii1 115 0 44 37 20 39 1.981199e-11 -8.824546e+00 -1.014823e+03
    utt> 17 chattiisagadxhq1 116 0 36 31 18 35 1.590836e-11 -9.825059e+00 -1.139707e+03
    utt> 18 daman1 56 0 28 22 13 25 6.779945e-12 -1.252179e+01 -7.012204e+02
    utt> 19 daadarnagarhawelii1 141 0 72 58 25 50 3.053612e-11 -1.066822e+01 -1.504218e+03
    utt> 20 damadamaa1 65 0 36 28 15 29 1.133063e-11 -1.149390e+01 -7.471032e+02
    utt> 21 delhii1 75 0 28 24 14 28 1.350518e-11 -7.321401e+00 -5.491051e+02
    utt> 22 dharmashaalaa1 102 0 40 34 18 36 1.899508e-11 -1.192620e+01 -1.216473e+03
    utt> 23 dilavaaraa1 68 0 36 28 15 30 1.296583e-11 -1.194053e+01 -8.119562e+02
    utt> 24 dillii1 75 0 24 21 13 25 5.892389e-12 -6.221243e+00 -4.665933e+02
    utt> 25 diuu1 61 0 20 17 11 21 4.373878e-12 -1.083887e+01 -6.611713e+02
    utt> 26 dxalalek1 92 0 32 27 16 32 1.380631e-11 -9.573455e+00 -8.807578e+02
    utt> 27 dalhauzi1 84 0 36 30 16 33 1.299172e-11 -7.204297e+00 -6.051609e+02
    utt> 28 dxiraamg1 87 0 32 27 16 31 1.294098e-11 -9.647626e+00 -8.393434e+02
    utt> 29 gir1 43 0 20 16 10 19 8.798983e-13 -1.005404e+01 -4.323238e+02
    utt> 30 girnaar1 75 0WARNING: "mk_phone_list.c", line 173: Unable to lookup (girnaar1) in the lexicon
    WARNING: "next_utt_states.c", line 82: Unable to produce CI phones for utt
    Tue Apr 23 13:47:08 2013

     
    • Pranav Jawale

      Pranav Jawale - 2013-04-25

      Please concentrate on the error messages.

      WARNING: "mk_phone_list.c", line 173: Unable to lookup (girnaar1) in the
      lexicon

      lexicon = dictionary

       
      • Pranav Jawale

        Pranav Jawale - 2013-04-25

        Share your transcription and filelist.

        WARNING: "corpus.c", line 1986: LSN utt id, amdhrpradesh1, does not match
        ctl utt id, aasaam1

         
        • Pranav Jawale

          Pranav Jawale - 2013-04-25

          Please read http://cmusphinx.sourceforge.net/wiki/tutorialam before
          preparing data for training. It says -

          It's critical to have exact match between fileids file and the
          transcription file. The number of lines in both should be identical. Last
          part of the file id (speaker1/file_1) and the utterance id file_1 must be
          the same on each line.

          In your case, fileids in transcription do not match with those in ctl file.

           
  • meshweta_7

    meshweta_7 - 2013-04-25

    here r the two files

     
  • meshweta_7

    meshweta_7 - 2013-04-25

    the transcription file

     
  • meshweta_7

    meshweta_7 - 2013-04-25

    This time again baum welch stops and the error is with girnaar1.
    Actually wav file is girnaar1.wav and in the transcript and dictionary "giranaar " is there; what could be the error. why it is in().

     
    • Nickolay V. Shmyrev

      You have extra space after symbol ")" in the line 31 with the word "girnaar". You need to remove this space to make training work.

      The log said that the word girnaar1 is missing, not the word girnaar. This is exactly what the problem is. Please read log carefully next time.

      This issue appears because you are using outdated Sphinxtrain. In new Sphinxtrain all those issues are bypassed. Please use latest released Sphinxtrain and latest updated tutorial from the website and read the logs.
      That will save both your time and our time.

       

Log in to post a comment.

Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.