Adaptation error - Total dimensionality of subvector specification > feature dimensionality

[Ruslan Selyukh]

I want to make adaptation eu_us generic model.
I tried and got an error.

bw.exe -hmmdir hmm/en_us -moddeffn hmm/en_us/mdef.txt -ts2cbfn .semi. -feat 1s_c_d_dd -lda hmm/en_us/feature_transform -svspec 0-12/13-25/26-38 -cmn current -agc none -dictfn cmu07a.dic -ctlfn ad.fileids -lsnfn ad.transcription -accumdir .

INFO: main.c(244): Reading linear feature transformation from hmm/en_us/feature_transform
INFO: main.c(254): Using subvector specification 0-12/13-25/26-38
ERROR: "feat.c", line 310: Total dimensionality of subvector specification 39 > feature dimensionality 32

Thanks in advance.

[Nickolay V. Shmyrev]

You need to remove -svspec option from the command line. -ts2cbfn must be .cont., not .semi.

[Bidisha Mukherjee]

Hi Nickolay, I had removed -svspec option and and instead of .semi. or .ptm. i used .cont. , after checking this i am getting the following error

INFO: feat.c(715): Initializing feature stream to type: '1s_c_d_dd', ceplen=13, CMN='batch', VARNORM='no', AGC='none'
INFO: acmod.c(152): Reading linear feature transformation from ./en-us/feature_transform
INFO: acmod.c(162): Using subvector specification 0-12/13-25/26-38
ERROR: "feat.c", line 311: Total dimensionality of subvector specification 39 > feature dimensionality 36

can you suggest any changes?

[Nickolay V. Shmyrev]

Tutorial says you need to add -lda

[Archil Gogorishvili]

Hi

I had this problem using Pocketsphinx-python. You can see code and full log here. Please also tell me how to remove -svspec from python config.
Code above is a python code part and log file.

MODELDIR = "/usr/local/share/pocketsphinx/model"
DATADIR = "/usr/local/share/pocketsphinx/test/data"

# Create a decoder with certain model
config = Decoder.default_config()
config.set_string('-hmm', path.join(MODELDIR, 'en-us/en-us/'))
config.set_string('-lm', path.join(MODELDIR, 'en-us/en-us.lm.bin'))
#config.set_string('-lm', path.join(MODELDIR, 'en-us/en-us-phone.lm.bin'))
config.set_string('-dict', path.join(MODELDIR, 'en-us/cmudict-en-us.dict'))
config.set_string('-logfn', path.join('/root','Desktop/recorder/flaskapi/flaskapi.log'))
#config.set_string('-inmic', 'yes')
decoder = Decoder(config)

INFO: pocketsphinx.c(152): Parsed model-specific feature parameters from /usr/local/share/pocketsphinx/model/en-us/en-us//feat.params
Current configuration:
[NAME]                  [DEFLT]         [VALUE]
-agc                    none            none
-agcthresh              2.0             2.000000e+00
-allphone
-allphone_ci            yes             yes
-alpha                  0.97            9.700000e-01
-ascale                 20.0            2.000000e+01
-aw                     1               1
-backtrace              no              no
-beam                   1e-48           1.000000e-48
-bestpath               yes             yes
-bestpathlw             9.5             9.500000e+00
-ceplen                 13              13
-cmn                    live            batch
-cmninit                40,3,-1         41.00,-5.29,-0.12,5.09,2.48,-4.07,-1.37,-1.78,-5.08,-2.05,-6.45,-1.42,1.17
-compallsen             no              no
-dict                                   /usr/local/share/pocketsphinx/model/en-us/cmudict-en-us.dict
-dictcase               no              no
-dither                 no              no
-doublebw               no              no
-ds                     1               1
-fdict
-feat                   1s_c_d_dd       1s_c_d_dd
-featparams
-fillprob               1e-8            1.000000e-08
-frate                  100             100
-fsg
-fsgusealtpron          yes             yes
-fsgusefiller           yes             yes
-fwdflat                yes             yes
-fwdflatbeam            1e-64           1.000000e-64
-fwdflatefwid           4               4
-fwdflatlw              8.5             8.500000e+00
-fwdflatsfwin           25              25
-fwdflatwbeam           7e-29           7.000000e-29
-fwdtree                yes             yes
-hmm                                    /usr/local/share/pocketsphinx/model/en-us/en-us/
-input_endian           little          little
-jsgf
-keyphrase
-kws
-kws_delay              10              10
-kws_plp                1e-1            1.000000e-01
-kws_threshold          1e-30           1.000000e-30
-latsize                5000            5000
-lda
-ldadim                 0               0
-lifter                 0               22
-lm                                     /usr/local/share/pocketsphinx/model/en-us/en-us.lm.bin
-lmctl
-lmname
-logbase                1.0001          1.000100e+00
-logfn                                  /root/Desktop/recorder/flaskapi/flaskapi.log
-logspec                no              no
-lowerf                 133.33334       1.300000e+02
-lpbeam                 1e-40           1.000000e-40
-lponlybeam             7e-29           7.000000e-29
-lw                     6.5             6.500000e+00
-maxhmmpf               30000           30000
-maxwpf                 -1              -1
-mdef
-mean
-mfclogdir
-min_endfr              0               0
-mixw
-mixwfloor              0.0000001       1.000000e-07
-mllr
-mmap                   yes             yes
-ncep                   13              13
-nfft                   512             512
-nfilt                  40              25
-nwpen                  1.0             1.000000e+00
-pbeam                  1e-48           1.000000e-48
-pip                    1.0             1.000000e+00
-pl_beam                1e-10           1.000000e-10
-pl_pbeam               1e-10           1.000000e-10
-pl_pip                 1.0             1.000000e+00
-pl_weight              3.0             3.000000e+00
-pl_window              5               5
-rawlogdir
-remove_dc              no              no
-remove_noise           yes             yes
-remove_silence         yes             yes
-round_filters          yes             yes
-samprate               16000           1.600000e+04
-seed                   -1              -1
-sendump
-senlogdir
-senmgau
-silprob                0.005           5.000000e-03
-smoothspec             no              no
-svspec                                 0-12/13-25/26-38
-tmat
-tmatfloor              0.0001          1.000000e-04
-topn                   4               4
-topn_beam              0               0
-toprule
-transform              legacy          dct
-unit_area              yes             yes
-upperf                 6855.4976       6.800000e+03
-uw                     1.0             1.000000e+00
-vad_postspeech         50              50
-vad_prespeech          20              20
-vad_startspeech        10              10
-vad_threshold          3.0             3.000000e+00
-var
-varfloor               0.0001          1.000000e-04
-varnorm                no              no
-verbose                no              no
-warp_params
-warp_type              inverse_linear  inverse_linear
-wbeam                  7e-29           7.000000e-29
-wip                    0.65            6.500000e-01
-wlen                   0.025625        2.562500e-02

INFO: feat.c(715): Initializing feature stream to type: '1s_c_d_dd', ceplen=13, CMN='batch', VARNORM='no', AGC='none'
INFO: acmod.c(152): Reading linear feature transformation from /usr/local/share/pocketsphinx/model/en-us/en-us//feature_transform
INFO: acmod.c(162): Using subvector specification 0-12/13-25/26-38
ERROR: "feat.c", line 311: Total dimensionality of subvector specification 39 > feature dimensionality 36

[Nickolay V. Shmyrev]

You mixed files from the different models in the folder /usr/local/share/pocketsphinx/model/en-us/en-us/, you need to clean the old model files before you unpack the new model.

[thierry LE CONNIAT]

i use the method describe in tutorial to adapt acoustic model for pocketsphinx and i observe that some audio file dont fail to align with audio and other fails.
My audio files contain many words like you see below :

<s> toulouse technique toulouse technique airbus one five six five five calling toulouse technique on one three two nine two how do it </s> (1)
<s> great cell call has    i call you on    on hash f one now </s> (4)
<s> toulouse thank you very much that's a good cell call check for air bus four five height charlie we're out thank you for your help bye bye </s> (8)

i try first command bw without -lda option and i have the first sentence fail. the others were ok
After i try bw with - lda option and result was only the first sentence OK.

Precision : I use a ptm model for the first command and a full model for second (because in ptm model there is no file feature_transform).

the first command that i use are this one (ptm model):

/bw -hmmdir en-us -moddeffn en-us/mdef.txt -ts2cbfn .ptm. -feat 1s_c_d_dd -svspec 0-12/13-25/26-38 -cmn current -agc none -dictfn cmudict-en-us.dict -ctlfn myfileids.fileids -lsnfn mytranscription.transcription -accumdir .

and the second with - lda option (full model) :

/bw -hmmdir en-us -moddeffn en-us/mdef.txt -ts2cbfn .cont. -feat 1s_c_d_dd -cmn current -agc none -dictfn cmudict-en-us.dict -ctlfn myfileids.fileids -lsnfn mytranscription.transcription
-lda en-us-adapt/feature_transform
-accumdir .

The error was :
ERROR: backward.c , line 421: Failed to align audio to trancript: final state of the search is not reached
ERROR: baum_welch.c, line 324: 4 ignored

What happens with my adaptation ?