Universiti Teknologi Malaysia Institutional Repository

Malay continuous speech recognition using fast HMM match algorithm

Ting, Chee-Ming and Salleh, Sh-Hussain and Ariff, A. K. (2009) Malay continuous speech recognition using fast HMM match algorithm. In: 2009 4th IEEE Conference on Industrial Electronics and Applications, ICIEA 2009. Institute of Electrical and Electronics Engineers, New York, pp. 2424-2427. ISBN 978-142442800-7

Full text not available from this repository.

Official URL: http://dx.doi.org/10.1109/ICIEA.2009.5138636

Abstract

This paper describes the implementation of fast hidden Markov model (HMM) match algorithm in a phoneme-based Malay continuous speech recognition system. The decoding algorithm decouples the computations of state-likelihoods of phone HMM's from the main search which is bounded by syntactical and lexical constraints. This avoids the redundant state-likelihood computations of identical phone HMM's for different word models in the tightly integrated search and thus substantially reduce the decoding time. The algorithm is implemented in the framework of one pass dynamic programming search. For a 541-word task, the fast HMM match reduces the real-time factor (RTF) by a factor of 31.8 from 286.45×RT to 9.02×RT, compared to without decoupling. The word accuracy is maintained at 91.6% without loss for a test set perplexity of 15.45 in speaker-dependent mode.

Item Type:Book Section
Additional Information:ISBN: 978-142442800-7; 2009 4th IEEE Conference on Industrial Electronics and Applications, ICIEA 2009; Xi'an; 25 May 2009 through 27 May 2009
Uncontrolled Keywords:acoustic match, continuous speech recognition, hidden markov model
Subjects:Q Science > Q Science (General)
Divisions:?? FBSK ??
ID Code:12940
Deposited By: Liza Porijo
Deposited On:07 Jul 2011 04:29
Last Modified:07 Jul 2011 04:29

Repository Staff Only: item control page