An efficient algorithm for parameterizing HsMM with Gaussian and Gamma distributions

被引：6

作者：

Xie, Y. ^{[1
]}

Tang, S. ^{[2
]}

Tang, C. ^{[3
]}

Huang, X. ^{[4
]}

机构：

[1] Sun Yat Sen Univ, Sch Informat Sci & Technol, Guangzhou 510275, Guangdong, Peoples R China

[2] Missouri Western State Univ, Dept Engn Technol, St Joseph, MO 64507 USA

[3] Guilin Univ Elect Technol, Sch Comp Sci & Engn, Guilin 541004, Peoples R China

[4] Sun Yat Sen Univ, Network & Informat Technol Ctr, Guangzhou 510275, Guangdong, Peoples R China

来源：

INFORMATION PROCESSING LETTERS | 2012年 / 112卷 / 19期

基金：

中国国家自然科学基金;

关键词：

Algorithm; Parametric HsMM; Gaussian; Gamma;

D O I：

10.1016/j.ipl.2012.06.005

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A widely used method for parameterizing hidden semi-Markov model is using Gaussian distribution to form the output probability and using Gamma distribution to form the state duration probability. Most of these models are based on the classical Newton's method with second-order convergence, whose iterative convergence speed is slow for large-scale realtime applications. An improved parameter re-estimation algorithm is introduced for such parametric hidden semi-Markov model in this paper. The proposed approach is based on forward and backward algorithm. It applies an iterative method with eighth-order convergence to improve the performance of the model. The numerical examples validate the proposed method. (C) 2012 Elsevier B.V. All rights reserved.

引用

页码：732 / 737

页数：6

共 10 条

[1] A new family of eighth-order iterative methods for solving nonlinear equations [J].

Bi, Weihong ;

Wu, Qingbiao ;

Ren, Hongmin .

APPLIED MATHEMATICS AND COMPUTATION, 2009, 214 (01) :236-245

[2] MAXIMUM LIKELIHOOD ESTIMATION OF PARAMETERS OF GAMMA DISTRIBUTION AND THEIR BIAS [J].

CHOI, SC ;

WETTE, R .

TECHNOMETRICS, 1969, 11 (04) :683-&

[3]

Ferguson J.D., 1980, S APPL HIDDEN MARKOV, P143

[4] CONTINUOUSLY VARIABLE DURATION HIDDEN MARKOV MODELS FOR AUTOMATIC SPEECH RECOGNITION. [J].

Levinson, S.E. .

Computer Speech and Language, 1986, 1 (01) :29-45

[5]

MCLACHLAN G, 2000, WILEY SER PROB STAT, P1, DOI 10.1002/0471721182

[6] A TUTORIAL ON HIDDEN MARKOV-MODELS AND SELECTED APPLICATIONS IN SPEECH RECOGNITION [J].

RABINER, LR .

PROCEEDINGS OF THE IEEE, 1989, 77 (02) :257-286

[7]

Russell M., 1993, ICASSP-93. 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing (Cat. No.92CH3252-4), P499, DOI 10.1109/ICASSP.1993.319351

[8]

Russell M. J., 1985, ICASSP 85. Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (Cat. No. 85CH2118-8), P5

[9] Hidden semi-Markov models [J].

Yu, Shun-Zheng .

ARTIFICIAL INTELLIGENCE, 2010, 174 (02) :215-243

[10] An efficient forward-backward algorithm for an explicit-duration hidden Markov model [J].

Yu, SZ ;

Kobayashi, H .

IEEE SIGNAL PROCESSING LETTERS, 2003, 10 (01) :11-14

← 1 →