8th Annual Conference of the International Speech Communication Association

Antwerp, Belgium
August 27-31, 2007

Linear and Non Linear Kernel GMM SuperVector Machines for Speaker Verification

Réda Dehak (1), Najim Dehak (2), Patrick Kenny (2), Pierre Dumouchel (2)

(1) LRDE, France
(2) CRIM, Canada

This paper presents a comparison between Support Vector Machines (SVM) speaker verification systems based on linear and non linear kernels defined in GMM supervector space. We describe how these kernel functions are related and we show how the nuisance attribute projection (NAP) technique can be used with both of these kernels to deal with the session variability problem. We demonstrate the importance of GMM model normalization (M-Norm) especially for the non linear kernel. All our experiments were performed on the core condition of NIST 2006 speaker recognition evaluation (all trials). Our best results (an equal error rate of 6.3%) were obtained using NAP and GMM model normalization with the non linear kernel.

Full Paper

Bibliographic reference.  Dehak, Réda / Dehak, Najim / Kenny, Patrick / Dumouchel, Pierre (2007): "Linear and non linear kernel GMM supervector machines for speaker verification", In INTERSPEECH-2007, 302-305.