Utterance Samples problem with SphinxTrain

[vkumar]

2010-08-11 18:19:02 IST
I am using SphinxTrain 1.0 from Sphinx release 3.5 on Fedora. I myself
recorded utterances in recording room with two microphone and followed the
steps given in the Robust Tutorial. While executing command: perl
scripts_pl/make_feats.pl -ctl etc/an4_train.fileids It gave following errors:

perl scripts_pl/make_feats.pl -ctl etc/an4_train.fileids -cfg not specified,

using the default ./etc/sphinx_train.cfg -param not specified, using the

default ./etc/feat.params bin/wave2feat \ -verbose yes \ -alpha 0.97 \ -dither

yes \ -doublebw no \ -nfilt 40 \ -ncep 13 \ -lowerf 133.33334 \ -upperf

6855.4976 \ -nfft 512 \ -wlen 0.0256 \ -c etc/an4_train.fileids \ -mswav yes \

-di /test_alfa8/an4/wav \ -ei wav \ -do /test_alfa8/an4/feat \ -eo mfc -help

no no -example no no -i -o -c etc/an4_train.fileids -nskip -runlen -di

/test_alfa8/an4/wav -ei wav -do /test_alfa8/an4/feat -eo mfc -nist no no -raw

no no -mswav no yes -input_endian little little -nchans 1 1 -whichchan 1 1

-logspec no no -feat sphinx sphinx -mach_endian little little -alpha 0.97

9.700000e-01 -srate 16000.0 1.600000e+04 -frate 100 100 -wlen 0.025625

2.560000e-02 -nfft 512 512 -nfilt 40 40 -lowerf 133.33334 1.333333e+02 -upperf

6855.4976 6.855498e+03 -ncep 13 13 -doublebw no no -warp_type inverse_linear

inverse_linear -warp_params -blocksize 200000 200000 -dither yes yes -seed -1

-1 -verbose no yes INFO: fe_interface.c(100): You are using the internal

mechanism to generate the seed. INFO: fe_sigproc.c(752): Current FE

Parameters: INFO: fe_sigproc.c(753): Sampling Rate: 16000.000000 INFO:

fe_sigproc.c(754): Frame Size: 410 INFO: fe_sigproc.c(755): Frame Shift: 160

INFO: fe_sigproc.c(756): FFT Size: 512 INFO: fe_sigproc.c(757): Lower

Frequency: 133.333 INFO: fe_sigproc.c(758): Upper Frequency: 6855.5 INFO:

fe_sigproc.c(759): Number of filters: 40 INFO: fe_sigproc.c(760): Number of

Overflow Samps: 0 INFO: fe_sigproc.c(761): Start Utt Status: 0 INFO:

fe_sigproc.c(763): Will add dither to audio INFO: fe_sigproc.c(764): Dither

seeded with -1 INFO: fe_sigproc.c(771): Will not use double bandwidth in mel

filter INFO: wave2feat.c(139): /test_alfa8/an4/wav/an4_clstk/sen1.wav LENGTH:

4 INFO: wave2feat.c(786): Reading MS Wav file

/test_alfa8/an4/wav/an4_clstk/sen1.wav: INFO: wave2feat.c(787): 16 bit PCM

data, 2 channels 169940 samples INFO: wave2feat.c(788): Sampled at 16000

ERROR: "wave2feat.c", line 883: unknown input file format ERROR:

"wave2feat.c", line 201: error reading speech data FATAL_ERROR: "wave2feat.c",

line 90: error converting files...exiting # The audio(utterances) has

following properties: Bit rate : 512 kbps Audio Sample size : 16 bits Channels

2 (stereo) Audio Sample rate : 16000 Hz Audio format: PCM I had previously
done Robust Tutorial successfully with utterances(.wav) recorded directly on
my LAPTOP (without microphone) on WINDOWS XP SP2 using Sound Recorder utility
installed with standard Windows XP. The thing i noted on doing this new run
is: INFO: wave2feat.c(787): 16 bit PCM data, 2 channels 169940 samples And
with previous run : INFO: wave2feat.c(787): 16 bit PCM data, 1 channels 28000
samples Does the errors with my new run is due to 2 channels(because of two
microphones used for recording) and large samples. Does SphinxTrain is capable
of handling certain samples and support only 1 channel. If above is not the
concern than what is to be, I am stuck at this time.

[Nickolay V. Shmyrev]

Does SphinxTrain is capable of handling certain samples

yes

and support only 1 channel.

yes