ISCA Archive Interspeech 2009
ISCA Archive Interspeech 2009

Fast GMM computation for speaker verification using scalar quantization and discrete densities

Guoli Ye, Brian Mak, Man-Wai Mak

Most of current state-of-the-art speaker verification (SV) systems use Gaussian mixture model (GMM) to represent the universal background model (UBM) and the speaker models (SM). For an SV system that employs log-likelihood ratio between SM and UBM to make the decision, its computational efficiency is largely determined by the GMM computation. This paper attempts to speedup GMM computation by converting a continuous-density GMM to a single or a mixture of discrete densities using scalar quantization. We investigated a spectrum of such discrete models: from high-density discrete models to discrete mixture models, and their combination called high-density discrete-mixture models. For the NIST 2002 SV task, we obtained an overall speedup by a factor of 2–100 with little loss in EER performance.


doi: 10.21437/Interspeech.2009-390

Cite as: Ye, G., Mak, B., Mak, M.-W. (2009) Fast GMM computation for speaker verification using scalar quantization and discrete densities. Proc. Interspeech 2009, 2327-2330, doi: 10.21437/Interspeech.2009-390

@inproceedings{ye09_interspeech,
  author={Guoli Ye and Brian Mak and Man-Wai Mak},
  title={{Fast GMM computation for speaker verification using scalar quantization and discrete densities}},
  year=2009,
  booktitle={Proc. Interspeech 2009},
  pages={2327--2330},
  doi={10.21437/Interspeech.2009-390}
}