DocumentCode :
1488910
Title :
The Student\´s 
-Hidden Markov Model With Truncated Stick-Breaking Priors
Author :
Wei, Xin ; Li, Chunguang
Author_Institution :
Dept. of Inf. Sci. & Electron. Eng., Zhejiang Univ., Hangzhou, China
Volume :
18
Issue :
6
fYear :
2011
fDate :
6/1/2011 12:00:00 AM
Firstpage :
355
Lastpage :
358
Abstract :
In this letter, we propose a Student´s t-hidden Markov model with truncated stick-breaking priors (TSB-SHMM). In the TSB-SHMM, the priors for elements in the initial state vector and the state transition matrix are constructed by stick-breaking procedure with a truncation level, and the observation emission distributions are the Student´s t-mixtures. Then we derive an inference algorithm for estimating the parameters of the proposed TSB-SHMM. Experimental results on the synthetic data and text-dependent speaker identification illustrate that the TSB-SHMM can automatically determine the number of states and are robust to untypical observed data.
Keywords :
hidden Markov models; inference mechanisms; matrix algebra; parameter estimation; speaker recognition; stochastic processes; user modelling; TSB-SHMM; inference algorithm; observation emission distribution; parameter estimation; state transition matrix; student t-hidden Markov model; synthetic data; text dependent speaker identification; truncated stick breaking prior; truncation level; Data models; Hidden Markov models; Inference algorithms; Markov processes; Probabilistic logic; Signal processing algorithms; Continuous hidden Markov model; Student\´s $t$ -distribution; inference; truncated stick-breaking priors;
fLanguage :
English
Journal_Title :
Signal Processing Letters, IEEE
Publisher :
ieee
ISSN :
1070-9908
Type :
jour
DOI :
10.1109/LSP.2011.2138695
Filename :
5742766
Link To Document :
