This paper proposes a prior distribution determination technique using cross validation for speech recognition based on the Bayesian approach. The Bayesian method is a statistical technique for estimating reliable predictive distributions by marginalizing model parameters and its approximate version, the variational Bayesian method has been applied to HMMbased speech recognition. Since prior distributions representing prior information about model parameters affect the posterior distributions and model selection, the determination of prior distributions is an important problem. However, it has not been thoroughly investigate in speech recognition. The proposed method can determine reliable prior distributions without tuning parameters and select an appropriate model structure dependently on the amount of training data. Continuous phoneme recognition experiments show that the proposed method achieved a higher performance than the conventional methods.
[1]
Jj Odell,et al.
The Use of Context in Large Vocabulary Speech Recognition
,
1995
.
[2]
S. J. Young,et al.
Tree-based state tying for high accuracy acoustic modelling
,
1994
.
[3]
Takahiro Shinozaki.
Hmm State Clustering Based on Efficient Cross-Validation
,
2006,
2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.
[4]
Naonori Ueda,et al.
Variational bayesian estimation and clustering for speech recognition
,
2004,
IEEE Transactions on Speech and Audio Processing.
[5]
Hagai Attias,et al.
Inferring Parameters and Structure of Latent Variable Models by Variational Bayes
,
1999,
UAI.