Large Margin Gaussian mixture models for speaker identification

Abstract : Gaussian mixture models (GMM) have been widely and successfully used in speaker recognition during the last decade. However, they are generally trained using the generative criterion of maximum likelihood estimation. In this paper, we propose a simple and efficient discriminative approach to learn GMM with a large margin criterion to solve the classification problem. Our approach is based on a recent work about the Large Margin GMM (LM-GMM) where each class is modeled by a mixture of ellipsoids and which has shown good results in speech recognition. We propose a simplification of the original algorithm and carry out preliminary experiments on a speaker identification task using NIST-SRE'2006 data. We compare the traditional generative GMM approach, the original LM-GMM one and our own version. The results suggest that our algorithm outperforms the two others.
Type de document :
Communication dans un congrès
Interspeech, Sep 2010, Makuhari, Japan. 2010
Liste complète des métadonnées

https://hal.inria.fr/inria-00532781
Contributeur : Khalid Daoudi <>
Soumis le : jeudi 4 novembre 2010 - 13:53:11
Dernière modification le : jeudi 11 janvier 2018 - 06:21:34

Identifiants

  • HAL Id : inria-00532781, version 1

Collections

Citation

Reda Jourani, Khalid Daoudi, Régine André-Obrecht, Driss Aboutajdine. Large Margin Gaussian mixture models for speaker identification. Interspeech, Sep 2010, Makuhari, Japan. 2010. 〈inria-00532781〉

Partager

Métriques

Consultations de la notice

261