vector matrix from extract_feature( )

abhijeet3922 / Speaker-identification-using-GMMs

It uses GMM to train a speaker identification model. The training and testing has been done on subset (34 speakers) from VoxForge data corpus.

https://appliedmachinelearning.wordpress.com/2017/11/14/spoken-speaker-identification-based-on-gaussian-mixture-models-python-implementation/

54 stars 37 forks source link

vector matrix from extract_feature( ) #1

Closed jumfina closed 6 years ago

jumfina commented 6 years ago

extract_feature( ) produces a matrix of vectors of dimensions 399 by 40. As I understand, 40 is the 20 MFCC+ 20 Delta MFCC features of a single speaker file. Could you say what is contained in the 399 dimension?

zsn900503 commented 6 years ago

399 should be your No. of frames. You basically separate the audio into small frames, and you got 20 MFCC + 20 DMFCC from each frame