Skip to topic | Skip to bottom
Home
Publications
Publications.200709-ODYSSEY-Ar1.3 - 04 Oct 2007 - 16:42 - RedaDehaktopic end

Start of topic | Skip to actions
Reda Dehak, Najim Dehak, Patrick Kenny, Pierre Dumouchel. Kernel Combination for SVM Speaker Verification. In the proceedings of The Speaker and Language Recognition Workshop (Odyssey 2008) Stellenbosch, South Africa January 21-24, 2008

We present a new approach for constructing the kernels used to build support vector machines for speaker verification. The idea is to construct new kernels by taking linear combination of many kernels such as the GLDS and GMM supervector kernels. In this new kernel combination, the combination weights are speaker dependent rather than universal weights on score level fusion and there is no need for extra-data to estimate them. An experiment on the NIST 2006 speaker recognition evaluation dataset (all trial) was done using three different kernel functions (GLDS kernel, linear and Gaussian GMM supervector kernels). We compared our kernel combination to the optimal linear score fusion obtained using logistic regression. This optimal score fusion was trained on the same test data. We had an equal error rate of $\simeq 5,9\%$ using the kernel combination technique which is better than the optimal score fusion system ($\simeq 6,0\%$).


to top

PublicationForm
Logo:  
Category:  
Title: Kernel Combination for SVM Speaker Verification
Authors: Reda Dehak, Najim Dehak, Patrick Kenny, Pierre Dumouchel
Type: InConference
Whereprefix: In the proceedings of
Where: The Speaker and Language Recognition Workshop (Odyssey 2008)
Ref:  
Place: Stellenbosch, South Africa
Date: January 21-24, 2008
Note:  
Lang: english
Keywords:  
Status: to appear


You are here: Publications > 200709-ODYSSEY-A

to top

Copyright © 1999-2010 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback