ISCA & IEEE Workshop on Spontaneous Speech Processing and Recognition

April 13-16, 2003
Tokyo Institute of Technology, Tokyo, Japan

Discriminative Training of GMM for Language Identification

Dan Qu, Bingxi Wang

Department of Information Science, Information Engineering University, Zhengzhou, China

In this paper, a discriminative training procedure for a Gaussian Mixture Model (GMM) language identification system is described. The proposal is based on the Generalized Probabilistic Descent (GPD) algorithm and Minimum Classification Error Rates formulated to estimate the GMM parameters. The evaluation is conducted using the OGI multi-language telephone speech corpus. The experimental results show such system is very effective in language identification tasks.


Full Paper

Bibliographic reference.  Qu, Dan / Wang, Bingxi (2003): "Discriminative training of GMM for language identification", in SSPR-2003, paper MAP8.