Welcome to the Feature extraction and voice matching application repository.
In this document, the codes used include two parts.
The first part is focused on extracting various sound features such as mfcc, zcr, chroma, energy_envelope
And in the second part, I have tried to use VGG-16 model to compare and match the input sound with a dataset of sample sounds.
- Higher code execution speed compared to the first code sample
- Better voice gender recognition
And I'm improving the error and optimizing the code.