The Definition and Extension of the Question Set for Decision Tree Based State Tying in Chinese Speech Recognition
作者： Fang Zheng , Jing Li , Wenhu Wu , Jiyong Zhang , Mingxing Xu
DOI:
关键词:
摘要: 1. Abstract This study deals with the decision tree based state tying method for acoustic modeling in Chinese Speech Recognition. In this paper, definition of context dependent Initial-Final units is given, and linguistic knowledge question set used described. The basic our experiment on classified contexts. Two methods extending refining are also proposed paper. One adding simple questions (corresponding to unclassified contexts) particular states after investigating influence contexts states. other one further two-side extended set. way, left right considered at same time during node’s splitting. experimental results show that two can improve performance model.
参考文章(8)
L.R. Bahl, P.V. deSouza, P.S. Gopalakrishnan, D. Nahamoo, M.A. Picheny, Decision trees for phonological rules in continuous speech [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing. pp. 185- 188 ,(1991) , 10.1109/ICASSP.1991.150308
L. Rabiner, B. Juang, An introduction to hidden Markov models IEEE ASSP Magazine. ,vol. 3, pp. 4- 16 ,(1986) , 10.1109/MASSP.1986.1165342
K. Beulen, H. Ney, Automatic question generation for decision tree based state tying international conference on acoustics speech and signal processing. ,vol. 2, pp. 805- 808 ,(1998) , 10.1109/ICASSP.1998.675387
W. Reichl, W. Chou, Decision tree state tying based on segmental clustering for acoustic modeling international conference on acoustics speech and signal processing. ,vol. 2, pp. 801- 804 ,(1998) , 10.1109/ICASSP.1998.675386
W. Reichl, Wu Chou, Robust decision tree state tying for continuous speech recognition IEEE Transactions on Speech and Audio Processing. ,vol. 8, pp. 555- 566 ,(2000) , 10.1109/89.861375
Kris Demuynck, Jacques Duchateau, Dirk Van Compernolle, A novel node splitting criterion in decision tree construction for semi-continuous HMMs. conference of the international speech communication association. ,vol. 3, pp. 1183- 1186 ,(1997)
Yinfei Huang, Wenhu Wu, Cheng Bi, Jian Wu, Mingxing Xu, Fang Zheng, Zhanjiang Song, EASYTALK: A LARGE-VOCABULARY SPEAKER-INDEPENDENT CHINESE DICTATION MACHINE conference of the international speech communication association. ,(1999)
Fang Zheng, Guoliang Zhang, Integrating the energy information into MFCC. conference of the international speech communication association. pp. 389- 392 ,(2000)