A structural Bayes approach to speaker adaptation.

Maximum a posteriori (MAP) estimation has been successfully applied to speaker adaptation in speech recognition systems using hidden Markov models. When the amount of data is sufficiently large, MAP estimation yields recognition performance as good as that obtained using maximum-likelihood (ML) esti...

Descrición completa

Detalles Bibliográficos
Publicado en:	IEEE Transactions on speech and audio processing 9, 3 (2001).
Autor Principal:	Shinoda, K.
Formato:	Artigo
Idioma:	English
Subjects:	MAP estimation. SMAP approach. Error rates. Hidden Markov models. Hierarchical structure. Maximum a posteriori estimation. Model parameter space. Nonnative speaker. Probability density functions. Speaker adaptation. Speech recognition systems. Structural Bayes approach. Structural maximum a posteriori approach. Supervised adaptation experiments. Unsupervised adaptation experiments. Utterances.