Title :
Hierarchical Correlation Compensation For Hidden Markov Models
Author :
Lin, Hui ; Tian, Ye ; Zhou, Jian-lai ; Jiang, Hui
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
fDate :
March 18-23, 2005
Keywords :
Gaussian distribution; compensation; correlation methods; covariance matrices; hidden Markov models; hierarchical systems; maximum likelihood estimation; speech recognition; trees (mathematics); CDHMM; Gaussian component leaf node representation; HCC; covariance space hierarchical tree; diagonal covariance matrices; hidden Markov models; hierarchical correlation compensation; linear combination weight estimation; maximum likelihood criterion; speech recognition; Asia; Computer science; Covariance matrix; Discrete cosine transforms; Hidden Markov models; Linear discriminant analysis; Maximum likelihood estimation; Prototypes; Reliability engineering; Speech recognition;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
Print_ISBN :
0-7803-8874-7
DOI :
10.1109/ICASSP.2005.1415203
