Sixth European Conference on Speech Communication and Technology

Budapest, Hungary
September 5-9, 1999

Combination of Vector Quantization and Gaussian Mixture Models for Speaker Verification with Sparse Training Data

Guido Kolano, Peter Regel-Brietzmann

DaimlerChrysler AG, Research and Technology, Ulm, Germany

We present a combination of an extended vector quantization (VQ) algorithm for training a speaker model and a gaussian interpretation of the VQ speaker model in the verification phase. This leads to a large decrease of the error rates compared to normal vector quantization and only a slight deterioration compared to full Gaussian mixture model (GMM) training. The training costs of the new method are only slightly higher than for pure vector quantization.

Full Paper (PDF)   Gnu-Zipped Postscript

Bibliographic reference.  Kolano, Guido / Regel-Brietzmann, Peter (1999): "Combination of vector quantization and gaussian mixture models for speaker verification with sparse training data", In EUROSPEECH'99, 1203-1206.