Menu

Getting the confidence of nbest hypotheses in pocketsphinx

Help
2015-09-06
2015-09-09
  • Yuval Karon

    Yuval Karon - 2015-09-06

    Hello,

    I would like to get the confidence score of the nbest hypothesis set.

    Using the Java interface through SWIG, Hypothesis.getProb() returns zero.
    On the other hand, getProb() returns a meaningful number
    with the hypothesis decoder.getHyp() . I am using the latest SVN version:13101 .

    Here is the configuration, both bestpath and fwdflat are enabled.

    Why do I get zero prob values on the nbest hypothesis set? is it because of
    the configuration, or the model training?

    Thanks,
    Yuval

    -agc none none
    -agcthresh 2.0 2.000000e+000
    -allphone
    -allphone_ci no no
    -alpha 0.97 9.700000e-001
    -ascale 20.0 2.000000e+001
    -aw 1 1
    -backtrace no no
    -beam 1e-48 1.000000e-048
    -bestpath yes yes
    -bestpathlw 9.5 9.500000e+000
    -ceplen 13 13
    -cmn current current
    -cmninit 8.0 8.0
    -compallsen no no
    -debug 0
    -dict C:\Users...\mft.dic
    -dictcase no no
    -dither no no
    -doublebw no no
    -ds 1 1
    -fdict C:\Users...\noisedict
    -feat 1s_c_d_dd 1s_c_d_dd
    -featparams C:\Users...\feat.params
    -fillprob 1e-8 1.000000e-008
    -frate 100 100
    -fsg
    -fsgusealtpron yes yes
    -fsgusefiller yes yes
    -fwdflat yes yes
    -fwdflatbeam 1e-64 1.000000e-064
    -fwdflatefwid 4 4
    -fwdflatlw 8.5 8.500000e+000
    -fwdflatsfwin 25 25
    -fwdflatwbeam 7e-29 7.000000e-029
    -fwdtree yes yes
    -hmm C:\Users...\cont_20
    -input_endian little little
    -jsgf
    -keyphrase
    -kws
    -kws_delay 10 10
    -kws_plp 1e-1 1.000000e-001
    -kws_threshold 1 1.000000e+000
    -latsize 5000 5000
    -lda
    -ldadim 0 0
    -lifter 0 0
    -lm C:\Users...\mft.lm.DMP
    -lmctl
    -lmname
    -logbase 1.0001 1.000100e+000
    -logfn
    -logspec no no
    -lowerf 133.33334 1.333334e+002
    -lpbeam 1e-40 1.000000e-040
    -lponlybeam 7e-29 7.000000e-029
    -lw 6.5 6.500000e+000
    -maxhmmpf 30000 30000
    -maxwpf -1 -1
    -mdef C:\Users...\mdef
    -mean C:\Users...\means
    -mfclogdir
    -min_endfr 0 0
    -mixw C:\Users...\mixture_weights
    -mixwfloor 0.0000001 1.000000e-007
    -mllr
    -mmap yes yes
    -ncep 13 13
    -nfft 512 512
    -nfilt 40 40
    -nwpen 1.0 1.000000e+000
    -pbeam 1e-48 1.000000e-048
    -pip 1.0 1.000000e+000
    -pl_beam 1e-10 1.000000e-010
    -pl_pbeam 1e-10 1.000000e-010
    -pl_pip 1.0 1.000000e+000
    -pl_weight 3.0 3.000000e+000
    -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+004
    -seed -1 -1
    -sendump
    -senlogdir
    -senmgau
    -silprob 0.005 5.000000e-003
    -smoothspec no no
    -svspec
    -tmat C:\Users...\transition_matrices
    -tmatfloor 0.0001 1.000000e-004
    -topn 4 4
    -topn_beam 0 0
    -toprule
    -transform legacy legacy
    -unit_area yes yes
    -upperf 6855.4976 6.855498e+003
    -uw 1.0 1.000000e+000
    -vad_postspeech 50 50
    -vad_prespeech 20 20
    -vad_startspeech 10 10
    -vad_threshold 2.0 2.000000e+000
    -var C:\Users...\variances
    -varfloor 0.0001 1.000000e-004
    -varnorm no no
    -verbose no no
    -warp_params
    -warp_type inverse_linear inverse_linear
    -wbeam 7e-29 7.000000e-029
    -wip 0.65 6.500000e-001
    -wlen 0.025625 2.562500e-002

     
    • Nickolay V. Shmyrev

      Hello Yuval

      We do not support this feature yet.

      Overall, the confidence of n-best hypothesis can be calculated with paths core, basically you compare prob of this particular n-best line with all other lines. This API has not been implemented yet though.

       
  • Yuval Karon

    Yuval Karon - 2015-09-09

    Thank you.

     

Log in to post a comment.