https://github.com/demfier/cmusphinx_training
Repo for cmusphinx activities
https://github.com/demfier/cmusphinx_training
Last synced: 4 months ago
JSON representation
Repo for cmusphinx activities
- Host: GitHub
- URL: https://github.com/demfier/cmusphinx_training
- Owner: Demfier
- License: other
- Created: 2017-03-12T00:08:40.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2020-05-17T15:24:02.000Z (about 6 years ago)
- Last Synced: 2025-07-09T14:54:14.730Z (11 months ago)
- Language: HTML
- Size: 79.1 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: Readme.md
- License: LICENSE
Awesome Lists containing this project
README
## Model Details
This model has been trained on the an4 database.
## Results
**SENTENCE ERROR** : 55.4%
**WORD ERROR RATE** : 18.5%
To see a more detailed description of the model result, check the [result folder](result/).
## Errors that I faced & how I countered them
I tried to follow the steps mentioned in the tutorial [here](http://cmusphinx.sourceforge.net/wiki/tutorialam) very closely, but I am a human my friend! Doing mistakes is in our nature! So don't worry if you too did the same mistakes I did, it's completely normal :)
Below is the list of errors I faced and how I counterd them.
**Note:** My first step was to check my log file (located at ```an4/logdir``` folder), google out the errors and see what comes. (Pretty obvious isn't it !)
* _Configuration (e.g. etc/sphinx_train.cfg) not defined_ :
Running the training
Configuration (e.g. etc/sphinx_train.cfg) not defined
Compilation failed in require at /usr/local/lib/sphinxtrain/scripts/000.comp_feat/slave_feat.pl line 51.
BEGIN failed--compilation aborted at /usr/local/lib/sphinxtrain/scripts/000.comp_feat/slave_feat.pl line 51.
**Solution:** Set environment variables using the following commands:
```export PATH=/usr/local/bin:$PATH```
```export LD_LIBRARY_PATH=/usr/local/lib```
```export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig```
* _Training failure while decoding the model_ :
Sphinxtrain path: /usr/local/lib/sphinxtrain
Sphinxtrain binaries path: /usr/local/libexec/sphinxtrain
Running the training
MODULE: 000 Computing feature from audio files
Extracting features from segments starting at (part 1 of 1)
Extracting features from segments starting at (part 1 of 1)
Feature extraction is done
MODULE: 00 verify training files
Phase 1: Checking to see if the dict and filler dict agrees with the phonelist file.
.
.
.
Skipped: $ST::CFG_MMIE set to 'no' in sphinx_train.cfg
MODULE: 65 MMIE Training
Skipped: $ST::CFG_MMIE set to 'no' in sphinx_train.cfg
MODULE: 90 deleted interpolation
Skipped for continuous models
MODULE: DECODE Decoding using models previously trained
Decoding 3 segments starting at 0 (part 1 of 1)
0%
This step had 3 ERROR messages and 0 WARNING messages. Please check the log file for details.
Aligning results to find error rate
word_align.pl failed with error code 65280 at /usr/local/lib/sphinxtrain/scripts/decode/slave.pl line 173.
**Solution:** Configure ```etc/sphinx_train.cfg``` properly: I believe this error arised due to some unset variables in our configuration file like.
* ${CFG_DIRLABEL}: Set to directory label such as ptm, ci etc..
* ${CFG_N_TIED_STATES}: Set to number of tied states value.
(e.g. $DEC_CFG_MODEL_NAME: Set $CFG_EXPTNAME.cd_${CFG_DIRLABEL}_${CFG_N_TIED_STATES} --> $CFG_EXPTNAME.cd_ptm_200)
* ${CFG_BASE_DIR}: Set to your database directory.
* ${CFG_EXPTNAME}: Set to your database name.
(e.g ${CFG_BASE_DIR}/model_architecture/${CFG_EXPTNAME}.tree_questions --> /home/gaurav/python-scripts/py/scripts/sphinx-source2/an4/model_architecture/an4.tree_questions)
* _Training failure while decoding the model_ :
**Solution**:
Well something similar came to my terminal like last time, but the reason was different. This time, it couldn't find the language model file. On searching my ```etc/``` folder, I found out that the our lm file was present by the name of ```an4.ug.lm.DMP```. Still not sure where that "ug" came from and I would love to know it. All I did was to change the filename to ```an4.lm.DMP``` and it seemed to do the trick.
## Additional Note:
Unfortunately, I won't be able to assist with the errors anymore as I am no longer maintaining this repository. You can, however, turn to these resources for help:
- [CMU Sphinx FAQ](https://cmusphinx.github.io/wiki/faq/)
- [CMU Sphinx Discussion Forum](https://sourceforge.net/p/cmusphinx/discussion/help/)
- [Sphinx4 Discussion Forum](https://sourceforge.net/p/cmusphinx/discussion/sphinx4/)
- Some other communication links can be found [here](https://cmusphinx.github.io/wiki/communicate/)