I want to find this feature —log mel filter bank energy for a speech data, and eventually feed it to the deep learning model.
Thats a different thing, as of now I am confused is it log of MFCC as we are doing in librosa or there is any other way?
any help will do
Log mel filterbank energy is not log of MFCC. This is computed prior to MFCC computation.
You can find these features implemented in python speech features package in Python:
import python_speech_features as psf
psf.logfbank()