Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

Dynamic Threshold Setting Via Bayesian Information Criterion (BIC) in HMM Training

Ying Jia, Yonghong Yan, Baosheng Yuan

Intel China Research Center, Beijing Kerry Center, Beijing, China

In this paper, an approach of dynamic threshold setting via Bayesian Information Criterion (BIC) in HMM training is described. The BIC threshold setting is applied to two important applications. Firstly, it is used to set the thresholds for decision tree based state tying, in place of the conventional approach of using a heuristic constant threshold. Secondly, it is applied to choosing the number of Gaussian mixture at state mixing-up stage. Experimental results on LVCSR Chinese dictation task indicate that BIC can dynamically set thresholds for cluster splitting according to the underlying complexity of the cluster parameters. Also significant performance improvement is achieved with the dynamic BIC threshold setting.


Full Paper

Bibliographic reference.  Jia, Ying / Yan, Yonghong / Yuan, Baosheng (2000): "Dynamic threshold setting via Bayesian information criterion (BIC) in HMM training", In ICSLP-2000, vol.4, 169-171.