explain the steps to extract mfcc features from voice signal using voice analysis tool box

I got the idea that mfcc features can be extracted from the voice signal using voice analysis tool box. Now i need the steps to install voice analysis tool box and in need of the mat lab code for the extraction of mfcc features. Need matlab code for the extraction of RPDE,DFA,MFCCs,WT based features and vocal fold features.

Answer 1

[Abhinav Aravindan]

Hi Akila,

I understand that you are looking for methods to extract features from a Voice Signal. To do this, you may use the “Audio Toolbox” in MATLAB. Please refer to the following link for instructions on installing additional toolboxes into an existing installation of MATLAB.

https://www.mathworks.com/matlabcentral/answers/101885

“audioFeatureExtractor” in the “Audio Toolbox” consists of multiple audio feature extractors, such as MFCCs which you may find useful.

https://www.mathworks.com/help/audio/ref/audiofeatureextractor.html

https://www.mathworks.com/help/audio/feature-extraction.html

You may further refer to the following documentation on “Wavelet Transforms” (in Wavelet Toolbox) and “Spectral Slope” which is directly related to the resonant characteristics of the vocal folds.

1. Wavelet Transforms: https://www.mathworks.com/discovery/wavelet-transforms.html
2. Spectral Slope: https://www.mathworks.com/help/audio/ug/spectraldescriptors.html#:~:text=characteristics%20of%20the-,vocal%20folds,-and%20has%20also

It appears that there aren’t any functions available in MATLAB to directly compute the RPDE (Recurrence period density entropy) and DFA (Detrended Fluctuation Analysis) of an audio signal, however, you may check out the following link on DFA which may be useful:

https://www.mathworks.com/matlabcentral/fileexchange/19795