Can I use .mfc files instead of .sph?

[rezaee]

Hi
When I use the tutorial method to get feature vectors from my wav failes, I get the .mfc file format. Can I use them to training acoustic model? how should change the sphinxtrain.cfg file?

Isn't it better to convert them to .sph format? If it is, how can I do that?

[Nickolay V. Shmyrev]

Can I use them to training acoustic model?

Yes

how should change the sphinxtrain.cfg file?

# Audio waveform and feature file information
$CFG_WAVFILES_DIR = "$CFG_BASE_DIR/wav";
$CFG_WAVFILE_EXTENSION = 'wav';
$CFG_WAVFILE_TYPE = 'mswav'; # one of nist, mswav, raw
$CFG_FEATFILES_DIR = "$CFG_BASE_DIR/feat";
$CFG_FEATFILE_EXTENSION = 'mfc';

Isn't it better to convert them to .sph format?

No

[rezaee]

May I ask how can I convert may recorded wav files to .mfc or .sph?
I think I need "feat.params" file that I don't have it!

[Nickolay V. Shmyrev]

wav files converted to mfc on the first step "comp_feats" when you run "sphinxtrain run".