Discriminative training of GMM based on Maximum Mutual Information for language identification
| Autor: | Qu Dan, Wang Bingxi, Dai Guannan, Yan Honggang |
|---|---|
| Rok vydání: | 2006 |
| Předmět: |
Conditional entropy
Language identification business.industry Estimation theory Speech recognition Computer Science::Computation and Language (Computational Linguistics and Natural Language and Speech Processing) Pattern recognition Mutual information Mixture model symbols.namesake Discriminative model Computer Science::Sound symbols Artificial intelligence business Random variable Gaussian process Mathematics |
| Zdroj: | 2006 6th World Congress on Intelligent Control and Automation. |
| DOI: | 10.1109/wcica.2006.1712616 |
| Popis: | In this paper, a discriminative training procedure based on maximum mutual information (MMI) for a Gaussian mixture model (GMM) language identification system is described. The idea is to find the model parameters lambda that minimize the conditional entropy Hlambda (C | X) of the random variable C given the random variable X , which means minimize the uncertainty in knowing what language was spoken given access to the utterance in X . The implementation of the proposal is based on the generalized probabilistic descent (GPD) algorithm 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 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|