Does SphinxTrain support training on raw audio files?
It seems there are options for this but the code is stubbed out. I understand
how to convert from raw to wav, but would prefer to operate directly on wav
files for training.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Does SphinxTrain support training on raw audio files?
Yes
It seems there are options for this but the code is stubbed
Not sure what code are you talking about. The input format is configured in
the beginning of etc/sphinx_train.cfg. Make sure your format is good for
feature extraction parameters configured in scripts_pl/make_feats.pl
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Does SphinxTrain support training on raw audio files?
It seems there are options for this but the code is stubbed out. I understand
how to convert from raw to wav, but would prefer to operate directly on wav
files for training.
Yes
Not sure what code are you talking about. The input format is configured in
the beginning of etc/sphinx_train.cfg. Make sure your format is good for
feature extraction parameters configured in scripts_pl/make_feats.pl