An Overview of SIDEKIT

SIDEKIT aims at providing the whole chain of tools required to perform speaker recognition.
The main tools available include:
  • Acoustic features extraction

    • Linear-Frequency Cepstral Coefficients (LFCC)
    • Mel-Frequency Cepstral Coefficients (MFCC)
    • RASTA filtering
    • Energy-based Voice Activity Detection (VAD)
    • normalization (CMS, CMVN, Short Term Gaussianization)
  • Modeling and classification

    • Gaussian Mixture Models (GMM)
    • i - vectors
    • Probabilistic Linear Discriminant Analysis (PLDA)
    • Joint Factor Analysis (JFA)
    • Support Vector Machine (SVM)
    • Deep Neural Network (bridge to THEANO)
  • Presentation of the results
    • DET plot
    • ROC Convex Hull based DET plot

Implementation

SIDEKIT has been designed and written in Python and released under LGPL License
to allow a wider usage of the code that, we hope, could be beneficial to the community.
The structure of the core package makes use of a limited number of classes in order
to facilitate the readability and reusability of the code.
Starting from version 1.1.0 SIDEKIT is no longer tested under Python 2.* In case you want to keep using Python2, you may have modification to do on your own.
SIDEKIT has been tested under Python >3.5 for both Linux and MacOS.

About SIDEKIT

Authors:Anthony Larcher & Kong Aik Lee & Sylvain Meignier
Version:1.2 of 2017/02/09

To know about the version and license of SIDEKIT

sidekit.__version__
sidekit.__license__