Menu ▾ ▴

problem loading -hmm model

Help
Pablo
2011-06-14
2012-09-22

  • Pablo

    Pablo - 2011-06-14

    Hi, i find this model: voxforge-en-0.4
    inside it has:
    etc
    model_parameters
    scripts
    readme
    problems

    inside the model_parameters
    voxforge_en_sphinx.cd_cont_5000

    when I load in my program the model like this: 

    config = cmd_ln_init(NULL, ps_args(), TRUE,
                         "-hmm", MODELDIR "/hmm/en_US/voxforge_en_sphinx.cd_cont_5000",
                           "-lm", "/usr/share/pocketsphinx/model/hoap/hoap3.lm" ,
                           "-dict", "/usr/share/pocketsphinx/model/hoap/hoap3.dic",
                           "-jsgf","/usr/share/pocketsphinx/grammar/HoapGrammar.jsgf",
                         NULL);

    compile ok and show me that: 

    pablo@pablo-desktop:~/Escritorio/reconContinuo/out/linux-x86$ ./reconContinuo 
INFO: cmd_ln.c(559): Parsing command line:
\
    -hmm /usr/local/share/pocketsphinx/model/hmm/en_US/voxforge_en_sphinx.cd_cont_5000 \
    -lm /usr/share/pocketsphinx/model/hoap/hoap3.lm \
    -dict /usr/share/pocketsphinx/model/hoap/hoap3.dic \
    -jsgf /usr/share/pocketsphinx/grammar/HoapGrammar.jsgf

Current configuration:
[NAME]      [DEFLT]     [VALUE]
-agc        none        none
-agcthresh  2.0     2.000000e+00
-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
-bghist     no      no
-ceplen     13      13
-cmn        current     current
-cmninit    8.0     8.0
-compallsen no      no
-debug              0
-dict               /usr/share/pocketsphinx/model/hoap/hoap3.dic
-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/hmm/en_US/voxforge_en_sphinx.cd_cont_5000
-input_endian   little      little
-jsgf               /usr/share/pocketsphinx/grammar/HoapGrammar.jsgf
-kdmaxbbi   -1      -1
-kdmaxdepth 0       0
-kdtree             
-latsize    5000        5000
-lda                
-ldadim     0       0
-lextreedump    0       0
-lifter     0       0
-lm             /usr/share/pocketsphinx/model/hoap/hoap3.lm
-lmctl              
-lmname     default     default
-logbase    1.0001      1.000100e+00
-logfn              
-logspec    no      no
-lowerf     133.33334   1.333333e+02
-lpbeam     1e-40       1.000000e-40
-lponlybeam 7e-29       7.000000e-29
-lw     6.5     6.500000e+00
-maxhmmpf   -1      -1
-maxnewoov  20      20
-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      40
-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-5        1.000000e-05
-pl_window  0       0
-rawlogdir          
-remove_dc  no      no
-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             
-tmat               
-tmatfloor  0.0001      1.000000e-04
-topn       4       4
-topn_beam  0       0
-toprule            
-transform  legacy      legacy
-unit_area  yes     yes
-upperf     6855.4976   6.855498e+03
-usewdphones    no      no
-uw     1.0     1.000000e+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(697): Initializing feature stream to type: '1s_c_d_dd', ceplen=13, CMN='current', VARNORM='no', AGC='none'
INFO: cmn.c(142): mean[0]= 12.00, mean[1..12]= 0.0
ERROR: "acmod.c", line 84: Acoustic model definition is not specified neither with -mdef option nor with -hmm
pablo@pablo-desktop:~/Escritorio/reconContinuo/out/linux-x86$

    I don't change anything, only put in my models and load. There must do
    something with the model? I read in the readme:

    To setup the files use scripts/build.sh script as a base. It should
    download required files from Voxforge, setup structure and extract
    features. Scripts are located in scripts subfolder.

    when I run $sh build.sh tell me that: 

    SOMETHING HERE
pablo@pablo-desktop:~/Escritorio/sphinxmodels/voxforge-en-0.4/scripts$ sh build.sh
find: «wav»: No existe el fichero o el directorio
build.sh: 64: ./scripts_pl/make_feats.pl: not found
build.sh: 65: ./scripts_pl/make_feats.pl: not found

    Next test in the reader:

    To use this model with pocketsphinx, just provide it's location with
    -hmm pocketsphinx option.
    Remember that this model uses MLLT and it requires both frontend
    modification. You need to use transform with LDA component and you need
    to adjust melfilter parameters.

    I don't know what's the meaning of the last paragraph. somebody tell me how I
    can load the language model? I suppose it is the largest model in english

     

  • Nickolay V. Shmyrev

    Nickolay V. Shmyrev - 2011-06-14

    ERROR: "acmod.c", line 84: Acoustic model definition is not specified
    neither with -mdef option nor with -hmm

    The model was missing in the path you specified in sources

    /usr/local/share/pocketsphinx/model/hmm/en_US/voxforge_en_sphinx.cd_cont_5000

    You need to specify model location properly

    pablo@pablo-desktop:~/Escritorio/sphinxmodels/voxforge-en-0.4/scripts$ sh
    build.sh
    find: «wav»: No existe el fichero o el directorio
    build.sh: 64: ./scripts_pl/make_feats.pl: not found
    build.sh: 65: ./scripts_pl/make_feats.pl: not fou

    You need to run this script from the root folder, not from the scripts folder.
    like this 

    ./script/build.sh

    You also need to edit this script to match your file layout. This script is
    not something you can just run, you need to edit it.

    I don't know what's the meaning of the last paragraph.

    This paragraph is applicable to sphinx4. You can learn more about sphinx4 on
    our website

    http://cmusphinx.sourceforge.net/sphinx4

    omebody tell me how I can load the language model?

    You just need to configure the decoder properly

    I suppose it is the largest model in english

    If you are about voxforge, it's not the largest model. The default model
    hub4_wsj_8k provided with pocketsphinx is better.

     

  • Anonymous

    Anonymous - 2011-07-28

    Hi!

    I'm having same problem here...

    I built an LM&dict using online LMtool and when I want to try it with both
    tidigits hmm or voxforge-en model, I get the same output you have.

    And the same if I use tidigits LM&dict with voxforge acoustic model.

    Any idea? Did you solve it?

     

  • Nickolay V. Shmyrev

    Nickolay V. Shmyrev - 2011-07-28

    Hello Ytniu

    I don't think your problem is related. Most likely the issue is that you are
    trying to combine tidigits dictionary with voxforge acoustic model. They are
    not compatible. You need to use tidigits acoustic model with tidigits
    dictionary.

    Next time please start a new thread. Please provide as precise description of
    the problem as possible. Include decoder logs. Describe command line
    parameters. Mention the version you are using. That will help you to resolve
    the issue quickly.

     

Log in to post a comment.