Hi, I'm using SphinxTrain to train a acoustic model. The environment is Ubuntu
11.10.
But I got this error:
terry@terry-VirtualBox:~/acoustic/alphas$ ./scripts_pl/RunAll.pl
MODULE: 00 verify training files
O.S. is case sensitive ("A" != "a").
Phones will be treated as case sensitive.
Phase 1: DICT - Checking to see if the dict and filler dict agrees with the
phonelist file.
Found 29 words using 55 phones
Phase 2: DICT - Checking to make sure there are not duplicate entries in the
dictionary
Phase 3: CTL - Check general format; utterance length (must be positive);
files exist
Phase 4: CTL - Checking number of lines in the transcript should match lines
in control file
Phase 5: CTL - Determine amount of training data, see if n_tied_states seems
reasonable.
Estimated Total Hours Training: 3.56006666666667
This is a small amount of data, no comment at this time
Phase 6: TRANSCRIPT - Checking that all the words in the transcript are in the
dictionary
Words in dictionary: 26
Words in filler dictionary: 3
Phase 7: TRANSCRIPT - Checking that all the phones in the transcript are in
the phonelist, and all phones in the phonelist appear at least once
Feature type is s2_4x which is 4 streams
LDA/MLLT only has sense for single stream features, for example 1s_c_d_dd
Skipping LDA training
Feature type is s2_4x which is 4 streams
LDA/MLLT only has sense for single stream features, for example 1s_c_d_dd
Skipping MLLT training
MODULE: 05 Vector Quantization
MODULE: 10 Training Context Independent models for forced alignment and VTLN
Skipped: $ST::CFG_FORCEDALIGN set to 'no' in sphinx_train.cfg
Skipped: $ST::CFG_VTLN set to 'no' in sphinx_train.cfg
MODULE: 11 Force-aligning transcripts
Skipped: $ST::CFG_FORCEDALIGN set to 'no' in sphinx_train.cfg
MODULE: 12 Force-aligning data for VTLN
Skipped: $ST::CFG_VTLN set to 'no' in sphinx_train.cfg
MODULE: 20 Training Context Independent models
Phase 1: Cleaning up directories:
accumulator...logs...qmanager...models...
Phase 2: Flat initialize
Phase 3: Forward-Backward
Baum welch starting for 1 Gaussian(s), iteration: 1 (1 of 1)
0%
This step had 1 ERROR messages and 0 WARNING messages. Please check the log
file for details.
Only 0 parts of 1 of Baum Welch were successfully completed
Parts 1 failed to run!
Training failed in iteration 1
Something failed:
(/home/terry/acoustic/alphas/scripts_pl/20.ci_hmm/slave_convg.pl)
I went to logdir and open alphas.1.1-1.bw.log. I saw this:
INFO: s3mixw_io.c(116): Read /home/terry/acoustic/alphas/model_parameters/alph
as.ci_semi_flatinitial/mixture_weights
INFO: s3tmat_io.c(115): Read /home/terry/acoustic/alphas/model_parameters/alph
as.ci_semi_flatinitial/transition_matrices
INFO: mod_inv.c(300): inserting tprob floor 1.000000e-04 and renormalizing
WARN: "s3io.c", line 256: Unable to open
/home/terry/acoustic/alphas/model_parameters/alphas.ci_semi_flatinitial/means
for reading; No such file or directory
FATAL_ERROR: "main.c", line 1905: initialization failed
It seems that there's no /alphas.ci_semi_flatinitial/means in
model_parameters/ folder. I also check the model_parameters/. It really
doesn't have this thing.
Why don't I have that? How can I solve this problem?
thank you!!!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
File 'means' is created by the tool bin/init_gau. It's log is located in the
file logdir/20.ci_hmm/an4.initmean_cihmm.log You can check that file for
details. You always need to check logs for details when you have issues. Not
just the last log but previous logs too.
I also have right environment variables
Most likely you don't
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi, I'm using SphinxTrain to train a acoustic model. The environment is Ubuntu
11.10.
But I got this error:
terry@terry-VirtualBox:~/acoustic/alphas$ ./scripts_pl/RunAll.pl
MODULE: 00 verify training files
O.S. is case sensitive ("A" != "a").
Phones will be treated as case sensitive.
Phase 1: DICT - Checking to see if the dict and filler dict agrees with the
phonelist file.
Found 29 words using 55 phones
Phase 2: DICT - Checking to make sure there are not duplicate entries in the
dictionary
Phase 3: CTL - Check general format; utterance length (must be positive);
files exist
Phase 4: CTL - Checking number of lines in the transcript should match lines
in control file
Phase 5: CTL - Determine amount of training data, see if n_tied_states seems
reasonable.
Estimated Total Hours Training: 3.56006666666667
This is a small amount of data, no comment at this time
Phase 6: TRANSCRIPT - Checking that all the words in the transcript are in the
dictionary
Words in dictionary: 26
Words in filler dictionary: 3
Phase 7: TRANSCRIPT - Checking that all the phones in the transcript are in
the phonelist, and all phones in the phonelist appear at least once
Feature type is s2_4x which is 4 streams
LDA/MLLT only has sense for single stream features, for example 1s_c_d_dd
Skipping LDA training
Feature type is s2_4x which is 4 streams
LDA/MLLT only has sense for single stream features, for example 1s_c_d_dd
Skipping MLLT training
MODULE: 05 Vector Quantization
MODULE: 10 Training Context Independent models for forced alignment and VTLN
Skipped: $ST::CFG_FORCEDALIGN set to 'no' in sphinx_train.cfg
Skipped: $ST::CFG_VTLN set to 'no' in sphinx_train.cfg
MODULE: 11 Force-aligning transcripts
Skipped: $ST::CFG_FORCEDALIGN set to 'no' in sphinx_train.cfg
MODULE: 12 Force-aligning data for VTLN
Skipped: $ST::CFG_VTLN set to 'no' in sphinx_train.cfg
MODULE: 20 Training Context Independent models
Phase 1: Cleaning up directories:
accumulator...logs...qmanager...models...
Phase 2: Flat initialize
Phase 3: Forward-Backward
Baum welch starting for 1 Gaussian(s), iteration: 1 (1 of 1)
0%
This step had 1 ERROR messages and 0 WARNING messages. Please check the log
file for details.
Only 0 parts of 1 of Baum Welch were successfully completed
Parts 1 failed to run!
Training failed in iteration 1
Something failed:
(/home/terry/acoustic/alphas/scripts_pl/20.ci_hmm/slave_convg.pl)
I went to logdir and open alphas.1.1-1.bw.log. I saw this:
INFO: s3mixw_io.c(116): Read /home/terry/acoustic/alphas/model_parameters/alph
as.ci_semi_flatinitial/mixture_weights
INFO: s3tmat_io.c(115): Read /home/terry/acoustic/alphas/model_parameters/alph
as.ci_semi_flatinitial/transition_matrices
INFO: mod_inv.c(300): inserting tprob floor 1.000000e-04 and renormalizing
WARN: "s3io.c", line 256: Unable to open
/home/terry/acoustic/alphas/model_parameters/alphas.ci_semi_flatinitial/means
for reading; No such file or directory
FATAL_ERROR: "main.c", line 1905: initialization failed
It seems that there's no /alphas.ci_semi_flatinitial/means in
model_parameters/ folder. I also check the model_parameters/. It really
doesn't have this thing.
Why don't I have that? How can I solve this problem?
thank you!!!
From http://cmusphinx.sourceforge.net/wiki/tutorialam
After you installed the software you may need to update the system
configuration so the system will be able to find the dynamic libraries. For
example
so you mean:
I have to setup these variables first. then do:
../SphinxTrain/scripts_pl/setup_SphinxTrain.pl -task an4
../pocketsphinx/scripts/setup_sphinx.pl -task an4
after that, I should have the correct model_parameters/ folder which contains
alphas.ci_semi_flatinitial/means. Is that right?
thank you!
SORRY, I was wrong.
I DO have alphas.ci_semi_flatinitial in model_parameters/ folder. But I don't
have "means" in alphas.ci_semi_flatinitial/ folder.
how can I get it? Is "means" made by the script?
by the way, I also have right environment variables. which are:
export PATH=/usr/local/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/lib
export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig
thank you!
File 'means' is created by the tool bin/init_gau. It's log is located in the
file logdir/20.ci_hmm/an4.initmean_cihmm.log You can check that file for
details. You always need to check logs for details when you have issues. Not
just the last log but previous logs too.
Most likely you don't
I have bin/init_gau, but I don't have that log file... I cannot read the
information in this log file.
why is that? thank you!
It means that you can't run this tool. Try to run it from a command line
without any arguments. See what happens.