How to overcome different MFCC coefficients distributions from different speech datasets?

1. How many MFCC coefficients are there?
2. How MFCC features are extracted for the speech recognition?
3. Why do we use DCT in MFCC?
4. What is frame blocking in MFCC?

How many MFCC coefficients are there?

2. There are 39 features of MFCC: a. 12 MFCC features.

How MFCC features are extracted for the speech recognition?

The MFCC feature extraction technique basically includes windowing the signal, applying the DFT, taking the log of the magnitude, and then warping the frequencies on a Mel scale, followed by applying the inverse DCT.

Why do we use DCT in MFCC?

DCT is the last step of the main process of MFCC feature extraction. The basic concept of DCT is correlating value of mel spectrum so as to produce a good representation of property spectral local. Basically the concept of DCT is the same as inverse fourier transform.

What is frame blocking in MFCC?

1) Frame Blocking: It means obtaining the characteristic features of the speech signal in a more stable state by processing it at small time intervals. 2) Windowing: Windowing can be defined as dividing a speech signal into a time period of a particular length.