CMU Sphinx / Forums / Help: Mic problems with pocketsphinx

Hi all,

after some work I managed to train a german acoustic model with SphinxTrain in
Windows XP. When testing it with scripts_pl/decode/slave.pl ->
pocketsphinx_batch.exe I get a correct word quota of about 66 %, not too good
but a beginning.

Now I'd like to do some live tests with pocketsphinx_continuous.exe and a
microphone. I tested with two different mics, one internal of my laptop, one
external via USB.

But whatever I do, not one recognized word is correct. When I get the prompt
"READY" from pocketsphinx_continous and I say some words, it runs through the
recognition routines. So I guess something from the mic gets through.

I experience the same problem when testing pocketsphinx_continuous with the
delivered english example models.

I made some tests with the command line arguments, e.g. -adcdev default and -input_endian little, but nothing worked. I have googled a lot but found only hints for parametrizing the mic input in Linux. Any ideas how to find the problem in Windows?
I pass the same -hmm, -lm and -dict parameters as in the successful pocketsphinx_batch tests. Is there something else I could have done wrong in that context?

Here is my version information:
-SphinxTrain 1.0
-SphinxBase 0.6
-PocketSphinx 0.6.0
-Visual Studio 2008

Here is one output from pocketsphinx_continuous.exe:

_C:\temp\tutorial\voxforge\bin>pocketsphinx_continuous
pocketsphinx_continuous_ar
gs.txt
INFO: cmd_ln.c(512): Parsing command line:
\
-hmm C:/temp/tutorial/voxforge/model_parameters/voxforge.cd_semi_1000 \
-dict C:/temp/tutorial/voxforge/etc/voxforge.dic \
-lm C:/temp/tutorial/voxforge/etc/voxforge.ug.lm

Current configuration:

-adcdev
-agc none none
-agcthresh 2.0 2.000000e+000
-alpha 0.97 9.700000e-001
-argfile
-ascale 20.0 2.000000e+001
-backtrace no no
-beam 1e-48 1.000000e-048
-bestpath yes yes
-bestpathlw 9.5 9.500000e+000
-bghist no no
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-compallsen no no
-debug 0
-dict C:/temp/tutorial/voxforge/etc/voxforge.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-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:/temp/tutorial/voxforge/model_parameters/voxfo
rge.cd_semi_1000
-input_endian little little
-jsgf
-kdmaxbbi -1 -1
-kdmaxdepth 0 0
-kdtree
-latsize 5000 5000
-lda
-ldadim 0 0
-lextreedump 0 0
-lifter 0 0
-lm C:/temp/tutorial/voxforge/etc/voxforge.ug.lm
-lmctl
-lmname default default
-logbase 1.0001 1.000100e+000
-logfn
-logspec no no
-lowerf 133.33334 1.333333e+002
-lpbeam 1e-40 1.000000e-040
-lponlybeam 7e-29 7.000000e-029
-lw 6.5 6.500000e+000
-maxhmmpf -1 -1
-maxnewoov 20 20
-maxwpf -1 -1
-mdef
-mean
-mfclogdir
-mixw
-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-5 1.000000e-005
-pl_window 0 0
-rawlogdir
-remove_dc no no
-round_filters yes yes
-samprate 16000 1.600000e+004
-seed -1 -1
-sendump
-senmgau
-silprob 0.005 5.000000e-003
-smoothspec no no
-svspec
-tmat
-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
-usewdphones no no
-uw 1.0 1.000000e+000
-var
-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

INFO: cmd_ln.c(512): Parsing command line:
\
-alpha 0.97 \
-dither yes \
-doublebw no \
-nfilt 40 \
-ncep 13 \
-lowerf 133.33334 \
-upperf 6855.4976 \
-nfft 512 \
-wlen 0.0256 \
-transform legacy \
-feat s2_4x \
-agc none \
-cmn current \
-varnorm no

Current configuration:

-agc none none
-agcthresh 2.0 2.000000e+000
-alpha 0.97 9.700000e-001
-ceplen 13 13
-cmn current current
-cmninit 8.0 8.0
-dither no yes
-doublebw no no
-feat 1s_c_d_dd s2_4x
-frate 100 100
-input_endian little little
-lda
-ldadim 0 0
-lifter 0 0
-logspec no no
-lowerf 133.33334 1.333333e+002
-ncep 13 13
-nfft 512 512
-nfilt 40 40
-remove_dc no no
-round_filters yes yes
-samprate 16000 1.600000e+004
-seed -1 -1
-smoothspec no no
-svspec
-transform legacy legacy
-unit_area yes yes
-upperf 6855.4976 6.855498e+003
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wlen 0.025625 2.560000e-002

INFO: acmod.c(238): Parsed model-specific feature parameters from
C:/temp/tutori
al/voxforge/model_parameters/voxforge.cd_semi_1000/feat.params
INFO: fe_interface.c(288): You are using the internal mechanism to generate
the
seed.
INFO: feat.c(848): Initializing feature stream to type: 's2_4x', ceplen=13,
CMN=
'current', VARNORM='no', AGC='none'
INFO: cmn.c(142): mean= 12.00, mean= 0.0
INFO: mdef.c(520): Reading model definition:
C:/temp/tutorial/voxforge/model_par
ameters/voxforge.cd_semi_1000/mdef
INFO: bin_mdef.c(173): Allocating 19470 * 8 bytes (152 KiB) for CD tree
INFO: tmat.c(205): Reading HMM transition probability matrices:
C:/temp/tutorial
/voxforge/model_parameters/voxforge.cd_semi_1000/transition_matrices
INFO: acmod.c(117): Attempting to use SCHMM computation module
INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
C:/temp/tutorial/vox
forge/model_parameters/voxforge.cd_semi_1000/means
INFO: ms_gauden.c(292): 1 codebook, 4 feature, size
256x12 256x24 256x3 256x12
INFO: ms_gauden.c(198): Reading mixture gaussian parameter:
C:/temp/tutorial/vox
forge/model_parameters/voxforge.cd_semi_1000/variances
INFO: ms_gauden.c(292): 1 codebook, 4 feature, size
256x12 256x24 256x3 256x12
INFO: ms_gauden.c(356): 135 variance values floored
INFO: s2_semi_mgau.c(897): Loading senones from dump file
C:/temp/tutorial/voxfo
rge/model_parameters/voxforge.cd_semi_1000/sendump
INFO: s2_semi_mgau.c(921): BEGIN FILE FORMAT DESCRIPTION
INFO: s2_semi_mgau.c(984): Rows: 256, Columns: 1208
INFO: s2_semi_mgau.c(1016): Using memory-mapped I/O for senones
INFO: s2_semi_mgau.c(1293): Maximum top-N: 4 Top-N beams: 0 0 0 0
INFO: dict.c(294): Allocating 7115 * 20 bytes (138 KiB) for word entries
INFO: dict.c(306): Reading main dictionary:
C:/temp/tutorial/voxforge/etc/voxfor
ge.dic
INFO: dict.c(206): Allocated 26 KiB for strings, 50 KiB for phones
INFO: dict.c(309): 3016 words read
INFO: dict.c(314): Reading filler dictionary:
C:/temp/tutorial/voxforge/model_pa
rameters/voxforge.cd_semi_1000/noisedict
INFO: dict.c(206): Allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(317): 3 words read
INFO: dict2pid.c(402): Building PID tables for dictionary
INFO: dict2pid.c(409): Allocating 3019 * 4 bytes (11 KiB) for word-internal
arra
ys
INFO: dict2pid.c(414): Allocating 41^3 * 2 bytes (134 KiB) for word-initial
trip
hones
INFO: dict2pid.c(453): Allocating 19833 entries of 2 bytes (38 KiB) for
internal
ssids
INFO: dict2pid.c(130): Allocated 20336 bytes (19 KiB) for word-final triphones
INFO: dict2pid.c(193): Allocated 20336 bytes (19 KiB) for single-phone word
trip
hones
INFO: ngram_model_arpa.c(476): ngrams 1=945, 2=2146, 3=2699
INFO: ngram_model_arpa.c(135): Reading unigrams
INFO: ngram_model_arpa.c(515): 945 = #unigrams created
INFO: ngram_model_arpa.c(194): Reading bigrams
INFO: ngram_model_arpa.c(531): 2146 = #bigrams created
INFO: ngram_model_arpa.c(532): 153 = #prob2 entries
INFO: ngram_model_arpa.c(539): 288 = #bo_wt2 entries
INFO: ngram_model_arpa.c(291): Reading trigrams
INFO: ngram_model_arpa.c(552): 2699 = #trigrams created
INFO: ngram_model_arpa.c(553): 77 = #prob3 entries
INFO: ngram_search_fwdtree.c(99): 239 unique initial diphones
INFO: ngram_search_fwdtree.c(147): 0 root, 0 non-root channels, 4 single-phone
w
ords
INFO: ngram_search_fwdtree.c(186): Creating search tree
INFO: ngram_search_fwdtree.c(191): before: 0 root, 0 non-root channels, 4
single
-phone words
INFO: ngram_search_fwdtree.c(324): after: max nonroot chan increased to 4218
INFO: ngram_search_fwdtree.c(333): after: 180 root, 4090 non-root channels, 3
si
ngle-phone words
INFO: ngram_search_fwdflat.c(153): fwdflat: min_ef_width = 4, max_sf_win = 25
Allocating 32 buffers of 2500 samples each
INFO: continuous.c(261): pocketsphinx_continuous COMPILED ON: May 5 2010, AT:
1
8:00:23

READY....
Listening...
Stopped listening, please wait...
INFO: cmn_prior.c(121): cmn_prior_update: from < 8.00 0.00 0.00 0.00 0.00
0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 >
INFO: cmn_prior.c(139): cmn_prior_update: to < 7.54 -0.43 0.18 0.42 0.10 -
0.24 0.00 -0.16 -0.00 -0.07 -0.26 -0.04 -0.12 >
INFO: ngram_search_fwdtree.c(1502): 3042 words recognized (37/fr)
INFO: ngram_search_fwdtree.c(1504): 70230 senones evaluated (846/fr)
INFO: ngram_search_fwdtree.c(1506): 182540 channels searched (2199/fr), 14179
1st, 50550 last
INFO: ngram_search_fwdtree.c(1510): 4879 words for which last channels evalu
ated (58/fr)
INFO: ngram_search_fwdtree.c(1513): 11173 candidate words for entering last p
hone (134/fr)
INFO: ngram_search_fwdflat.c(295): Utterance vocabulary contains 160 words
INFO: ngram_search_fwdflat.c(912): 2174 words recognized (26/fr)
INFO: ngram_search_fwdflat.c(914): 60620 senones evaluated (730/fr)
INFO: ngram_search_fwdflat.c(916): 148514 channels searched (1789/fr)
INFO: ngram_search_fwdflat.c(918): 11393 words searched (137/fr)
INFO: ngram_search_fwdflat.c(920): 7724 word transitions (93/fr)
WARNING: "ngram_search.c", line 1082: not found in last frame, using
<sil>
instead
INFO: ngram_search.c(1132): lattice start node .0 end node <sil>.62
INFO: ps_lattice.c(1228): Normalizer P(O) = alpha(<sil>:62:81) = -906817
INFO: ps_lattice.c(1266): Joint P(O,S) = -927457 P(S|O) = -20640
000000000: JUNI (-17379361)
READY...._ </sil></sil></sil>

~~Any help will be appreciated, thanks!~~

Kind regards
Michael

Mic problems with pocketsphinx_continuous Win

Speech Recognition Toolkit

Forums

Help

Mic problems with pocketsphinx_continuous Win

Mic problems with pocketsphinx_continuous Win

Speech Recognition Toolkit

Forums

Help

Mic problems with pocketsphinx_continuous Win document.SUBSCRIPTION_OPTIONS = { "thing": "topic", "subscribed": false, "url": "subscribe", "icon": { "css": "fa fa-envelope-o" } };

Mic problems with pocketsphinx_continuous Win