This project attempts to provide the functionality offered by various libraries used for speech/audio processing in a 'all-in-one' fashion. Here, libraries like bob.kaldi (python wrapper for the Kaldi Project), librosa, python_speech_features, cyvlfeat, kaldi.io among others are utilized.
The main purpose of this code is feature extraction. Which can be done easily in a Kaldi's "style". See "recipes" folder. Features that can be extracted are: mffcs, plp, i-vectors, fisher vectors. To be added in the future: filter banks, x-vectors.
Readme is going to be updated with more details in the future.
Note: this project is on its early stages and it's used in a pesonal manner only. There's no plans yet to give some kind of support and is currently under development and enhancement. However, any suggestion or pull request is be welcome.