Publication Details

Support vector machines and joint factor analysis for speaker verification

DEHAK, N.; KENNY, P.; DEHAK, R.; GLEMBEK, O.; DUMOUCHEL, P.; BURGET, L.; HUBEIKA, V.; CASTALDO, F. Support vector machines and joint factor analysis for speaker verification. Proc. ICASSP 2009. Taiwan: IEEE Signal Processing Society, 2009. p. 1-4. ISBN: 978-1-4244-2354-5.
Czech title
Support vector machines a joint factor analysis pro verifikaci mluvčího
Type
conference paper
Language
English
Authors
Dehak Najim
Kenny Patrick
Dehak Reda
Glembek Ondřej, Ing., Ph.D.
Dumouchel Pierre
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
Hubeika Valiantsina, Ing.
Castaldo Fabio
URL
Keywords

Joint Factor Analysis, Support Vector Machine, Speaker factors space, Within Class Covariance Normalization

Abstract

The paper is on support vector machines and joint factor analysis. Several variants of joint factor analysis are tested.

Annotation

This article presents several techniques to combine between Support vector machines (SVM) and Joint Factor Analysis (JFA) model for speaker verification. In this combination, the SVMs are applied in different sources of information produced by the JFA. These informations are the Gaussian Mixture Model supervectors and speakers and Common factors. We found that the use of JFA factors gave the best results especially when within class covariance normalization method is applied in the speaker factors space in order to compensate for the channel effect. The new combination results are comparable to other classical JFA scoring techniques.

Published
2009
Pages
1–4
Proceedings
Proc. ICASSP 2009
ISBN
978-1-4244-2354-5
Publisher
IEEE Signal Processing Society
Place
Taiwan
BibTeX 
@inproceedings{BUT33739,
  author="Najim {Dehak} and Patrick {Kenny} and Reda {Dehak} and Ondřej {Glembek} and Pierre {Dumouchel} and Lukáš {Burget} and Valiantsina {Hubeika} and Fabio {Castaldo}",
  title="Support vector machines and joint factor analysis for speaker verification",
  booktitle="Proc. ICASSP 2009",
  year="2009",
  pages="1--4",
  publisher="IEEE Signal Processing Society",
  address="Taiwan",
  isbn="978-1-4244-2354-5",
  url="http://www.fit.vutbr.cz/research/groups/speech/publi/2009/NAJIM_ICASSP2009.pdf"
}
Back to top