Enhanced time duration constraints in hidden Markov modelling for phoneme recognition

22 June 1989

journal article
Published by Institution of Engineering and Technology (IET) in Electronics Letters

Vol. 25 (13) , 824-825
https://doi.org/10.1049/el:19890555

Abstract

The use of enhanced time duration constraints for subword (phoneme) recognition in continuous speech is reported. Here the time duration constraints are modelled by a Gaussian probability distribution in the conventional Baum-Welch learning algorithm and are statistically enhanced to obtain the most probable path in the Viterbi decoding process. Experimental results to validate this approach are included.

Keywords

SPEECH RECOGNITION
CONTINUOUS SPEECH
PHONEME RECOGNITION
STATISTICALLY ENHANCED
VITERBI DECODING PROCESS
CONVENTIONAL BAUM-WELCH LEARNING ALGORITHM
ENHANCED TIME DURATION CONSTRAINTS
GAUSSIAN PROBABILITY DISTRIBUTION
HIDDEN MARKOV MODELLING

This publication has 0 references indexed in Scilit: