SPEAR: A Speaker Recognition Toolkit based on Bob

This package is part of the bob.bio packages, which provide open source tools to run comparable and reproducible biometric recognition experiments. In this package, tools for executing speaker recognition experiments are provided.

This includes:

• Speaker Recognition databases and their according protocols
• Voice activity detection
• Feature extraction
• Recognition/Verification tools

Notice that most of the machine learning tools (GMM, ISV, JFA, IVectors) are now handled by bob.bio.gmm.

spear is adapted to run speaker verification/recognition experiments with the SGE grid infrastructure at Idiap.

If you use this package and/or its results, please cite the following paper published at ICASSP 2014:

@inproceedings{spear,
  author = {Khoury, E. and El Shafey, L. and Marcel, S.},
  title = {Spear: An open source toolbox for speaker recognition based on {B}ob},
  booktitle = {IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP)},
  year = {2014},
  url = {http://publications.idiap.ch/downloads/papers/2014/Khoury_ICASSP_2014.pdf},
}

For more detailed information about the structure of the bob.bio packages, please refer to the documentation of bob.bio.base. Particularly, the installation of this and other bob.bio packages, please read the Installation Instructions.

In the following, we provide more detailed information about the particularities of this package only.

Users Guide

• Executing Baseline Algorithms
  • Setting up your Database
  • Running Baseline Experiments
  • The Algorithms
  • Evaluation Results
  • Experiments on different databases
• Implementation Details
  • 1. Voice Activity Detection
  • 2. Feature Extraction
  • 3. Universal Background Model Training
  • 4. Subspace Training
  • 5. Conditioning and Compensation
  • 6. Projection
  • 6. Model Enrollment
  • 7. Scoring
  • 8. Fusion
  • 9. Evaluation
• References

Reference Manual

• Tools implemented in bob.bio.spear
  • Summary
  • Databases
  • Preprocessors
  • Extractors