I'm trying to use livepretend, it correctly guesess about 20% of the words, but after about 20 seconds into the audio I get hundreds of messsages...
ERROR: "vithist.c", line 564: No word exits from last frame in block 981
INFO: main_live_pretend.c(92): PARTIAL HYP:
INFO: live.c(239): live_nfeatvec: 13
and it stops recognizing most of the words after this... What could cause this?
I'm trying to use livepretend, it correctly guesess about 20% of the words, but after about 20 seconds into the audio I get hundreds of messsages...
ERROR: "vithist.c", line 564: No word exits from last frame in block 981
INFO: main_live_pretend.c(92): PARTIAL HYP:
INFO: live.c(239): live_nfeatvec: 13
and it stops recognizing most of the words after this... What could cause this?
Here are my args:
INFO: cmd_ln.c(276): Parsing command line:
\
-mdef hub4opensrc.6000.mdef \
-dict cmudict.06d \
-fdict fillerdict \
-lm language_model.arpaformat.DMP \
-mean hub4opensrc_jan2002/6000senones/1s_c_d_dd/hub4opensrc.cd_continuous_8gau/means \
-var hub4opensrc_jan2002/6000senones/1s_c_d_dd/hub4opensrc.cd_continuous_8gau/variances \
-mixw hub4opensrc_jan2002/6000senones/1s_c_d_dd/hub4opensrc.cd_continuous_8gau/mixture_weights \
-tmat hub4opensrc_jan2002/6000senones/1s_c_d_dd/hub4opensrc.cd_continuous_8gau/transition_matrices \
-subvq hub4opensrc_jan2002/6000senones/1s_c_d_dd/hub4opensrc.cd_continuous_8gau/8gau.6000sen.quant \
-lowerf 133.33334 \
-upperf 6855.49756 \
-nfilt 40 \
-agc none \
-samprate 16000 \
-subvqbeam 1e-02 \
-epl 4 \
-fillprob 0.02 \
-lw 9.5 \
-maxwpf 10 \
-wip 0.00000000000001 \
-lextreedump 0 \
-cmn current