Subspace Distribution Clustering HMM for Chinese Digit Speech Recognition  

作　　者：秦伟 韦岗 

机构地区：[1]College of Electronics and Communications, South China University of Technology Guangzhou 510640 China

出　　处：《Journal of Electronic Science and Technology of China》2006年第1期43-46,共4页中国电子科技（英文版）

基　　金：Supported by the National Natural Science Foundation of China (No.60172048)

摘　　要：As a kind of statistical method, the technique of Hidden Markov Model （HMM） is widely used for speech recognition. In order to train the HMM to be more effective with much less amount of data, the Subspace Distribution Clustering Hidden Markov Model （SDCHMM）, derived from the Continuous Density Hidden Markov Model （CDHMM）, is introduced. With parameter tying, a new method to train SDCHMMs is described. Compared with the conventional training method, an SDCHMM recognizer trained by means of the new method achieves higher accuracy and speed. Experiment results show that the SDCHMM recognizer outperforms the CDHMM recognizer on speech recognition of Chinese digits.As a kind of statistical method, the technique of Hidden Markov Model （HMM） is widely used for speech recognition. In order to train the HMM to be more effective with much less amount of data, the Subspace Distribution Clustering Hidden Markov Model （SDCHMM）, derived from the Continuous Density Hidden Markov Model （CDHMM）, is introduced. With parameter tying, a new method to train SDCHMMs is described. Compared with the conventional training method, an SDCHMM recognizer trained by means of the new method achieves higher accuracy and speed. Experiment results show that the SDCHMM recognizer outperforms the CDHMM recognizer on speech recognition of Chinese digits.

关 键 词：speech recognition Subspace Distribution Clustering Hidden Markov Model（SDCHMM） Continuous Density Hidden Markov Model （CDHMM） parameter tying 

分 类 号：TN912.34[电子电信—通信与信息系统]