Menu

problem in training syllable models

Help
2009-03-25
2012-09-22
  • suresh chandra sekaran

          Iam working on speech recognition using syllable models. when i was doing syllable models for a small corpus which had around 100 syllables there was no problem, everything worked well but when i extended it for a large corpus having 479 syllables , sphinx is showing the following error.     "FATAL_ERROR: "ckd_alloc.c", line 79: Calloc failed from itree.c(64) ".
    

    I dont know what to do. should i have to make any correction in the configuration file due to the large no of syllables present. please help me....

    The detailed output is showm below.

    O.S. is case sensitive ("A" != "a").
    Phones will be treated as case sensitive.
    MODULE: 00 verify training files
    Phase 1: DICT - Checking to see if the dict and filler dict agrees with the phonelist file
    Found 921 words using 479 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.
    Total Hours Training: 1.45056388888889
    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: 917
    Words in filler dictionary: 2
    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
    MODULE: 20 Training Context Independent models
    Skipped: $ST::CFG_FORCEDALIGN set to 'no' in sphinx_train.cfg
    MODULE: 03 Force-aligning transcripts
    Skipped: $ST::CFG_FORCEDALIGN set to 'no' in sphinx_train.cfg
    MODULE: 10 Vector Quantization
    Skipped for continuous models

    MODULE: 20 Training Context Independent models
    Cleaning up directories: accumulator...logs...qmanager...models...
    Flat initialize
    Baum welch starting for 1 Gaussian(s), iteration: 1 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 0 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 1 Current Overall Likelihood Per Frame = -8.43250075164068
    Baum welch starting for 1 Gaussian(s), iteration: 2 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 54 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 2 Current Overall Likelihood Per Frame = -7.19394634948299
    Convergence Ratio = 0.146878658969192
    Baum welch starting for 1 Gaussian(s), iteration: 3 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 14 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 3 Current Overall Likelihood Per Frame = -4.27766077921046
    Convergence Ratio = 0.405380500298298
    Baum welch starting for 1 Gaussian(s), iteration: 4 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 6 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 4 Current Overall Likelihood Per Frame = -2.13939893664604
    Convergence Ratio = 0.4998670892644
    Baum welch starting for 1 Gaussian(s), iteration: 5 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 14 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 5 Current Overall Likelihood Per Frame = -1.14601645146817
    Convergence Ratio = 0.464327839077647
    Baum welch starting for 1 Gaussian(s), iteration: 6 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 14 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 6 Current Overall Likelihood Per Frame = -0.312890285471667
    Convergence Ratio = 0.726975747101343
    Baum welch starting for 1 Gaussian(s), iteration: 7 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 10 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 7 Current Overall Likelihood Per Frame = -0.0346142819187958
    Convergence Ratio = 0.889372462086458
    Baum welch starting for 1 Gaussian(s), iteration: 8 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 10 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 8 Current Overall Likelihood Per Frame = 0.261901466815238
    Convergence Ratio = 8.56628340375953
    Baum welch starting for 1 Gaussian(s), iteration: 9 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 10 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 9 Current Overall Likelihood Per Frame = 0.470972999337772
    Convergence Ratio = 0.798283167577776
    Baum welch starting for 1 Gaussian(s), iteration: 10 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 12 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 10 Current Overall Likelihood Per Frame = 0.592381507063801
    Convergence Ratio = 0.257782310019341
    Baum welch starting for 1 Gaussian(s), iteration: 11 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 10 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 11 Current Overall Likelihood Per Frame = 0.80444230060657
    Convergence Ratio = 0.357980104061434
    Baum welch starting for 1 Gaussian(s), iteration: 12 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 14 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 12 Current Overall Likelihood Per Frame = 0.878215266004682
    Convergence Ratio = 0.0917069693407291
    Baum welch starting for 1 Gaussian(s), iteration: 13 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 16 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 13 Current Overall Likelihood Per Frame = 0.93214657803454
    Convergence Ratio = 0.0614101281513943
    Baum welch starting for 1 Gaussian(s), iteration: 14 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 18 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 14 Current Overall Likelihood Per Frame = 1.03341257410766
    Convergence Ratio = 0.108637416538767
    Baum welch starting for 1 Gaussian(s), iteration: 15 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 20 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 15 Current Overall Likelihood Per Frame = 1.13448643273192
    Convergence Ratio = 0.0978059113626888
    Baum welch starting for 1 Gaussian(s), iteration: 16 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 20 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 16 Current Overall Likelihood Per Frame = 1.18052844237567
    Convergence Ratio = 0.040584010804677
    Baum welch starting for 1 Gaussian(s), iteration: 17 (1 of 1) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
    This step had 20 ERROR messages and 2 WARNING messages.
    Please check the log file for details. Normalization for iteration: 17 Current Overall Likelihood Per Frame = 1.22419586672572
    Training completed after 17 iterations
    MODULE: 30 Training Context Dependent models
    Cleaning up directories: accumulator...logs...qmanager...
    Initialization Make Untied Mdef
    FATAL_ERROR: "ckd_alloc.c", line 79: Calloc failed from itree.c(64)

                This step had 0 ERROR messages and 1 WARNING messages.
                Please check the log file for details.Something failed: (/home/project_student/Desktop/subbu/rm/scripts_pl/30.cd_hmm_untied/slave_convg.pl)
    
     
    • Nickolay V. Shmyrev

      it just fails to allocate required memoty for triphones which are trisyllables in your case. actually you just dont need to proceed further then ci training, so you dont need this step. just be sure you arr using enough (7 or more) states per hmm.

       

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.