the "pocketsphinx_batch" binary file requires some parameters like : -alpha
-doublebw nfilt ncep -lowerf -upperf -nfft -wlen -samprate -agc , etc. These
parameters are related to feature extraction. So I cannot understand what the
effect of these parameters on decoding as long as the decoder deals directly
with the feature vectors and not the audio files?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
the "pocketsphinx_batch" binary file requires some parameters like : -alpha
-doublebw nfilt ncep -lowerf -upperf -nfft -wlen -samprate -agc , etc. These
parameters are related to feature extraction. So I cannot understand what the
effect of these parameters on decoding as long as the decoder deals directly
with the feature vectors and not the audio files?
Those could be used if you decode wav files with -adcin yes. Otherwise they
are ignored
Agc is not related to feature extraction. It's applied on the feature stream.
So -agc and -cmn are used in each run.