Offline and online identification of hidden semi-Markov models

Author

Azimi, Mehran ; Nasiopoulos, Panos ; Ward, Rabab Kreidieh

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of British Columbia, Vancouver, BC, Canada

Volume

53

Issue

8

fYear

2005

Firstpage

2658

Lastpage

2663

Abstract

We present a new signal model for hidden semi-Markov models (HSMMs). Instead of constant transition probabilities used in existing models, we use state-duration-dependant transition probabilities. We show that our modeling approach leads to easy and efficient implementation of parameter identification algorithms. Then, we present a variant of the EM algorithm and an adaptive algorithm for parameter identification of HSMMs in the offline and online cases, respectively.

Keywords

hidden Markov models; probability; recursive estimation; signal processing; adaptive algorithm; expectation maximization algorithm; hidden; offline identification; online identification; parameter identification algorithm; recursive error prediction; recursive maximum likelihood; semiMarkov model; state-duration-dependant transition probability; Adaptive algorithm; Hidden Markov models; Maximum likelihood estimation; Parameter estimation; Power engineering and energy; Predictive models; Signal processing; Speech processing; State estimation; Tensile stress; Expectation maximization (EM) algorithm; recursive maximum likelihood (RML); recursive prediction error (RPE); semi-Markov models;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/TSP.2005.850344

Filename

1468462