Title :
Fuzzy evolutionary programming for hidden Markov modelling in speaker identification
Author :
Van Le, Tu ; Tran, Dat ; Wagner, Michael
Author_Institution :
Sch. of Comput., Canberra Univ., Belconnen, ACT, Australia
Abstract :
Gibbs distribution is used to represent fuzzy codebooks of individual speakers. The method of fuzzy evolutionary programming is employed to create the fuzzy codebooks and also to train hidden Markov models of speakers. This method increases the chance of attaining global maxima in the Baum-Welch algorithm for hidden Markov model re-estimation. The experiments show the results of speaker identification are very encouraging
Keywords :
evolutionary computation; fuzzy set theory; hidden Markov models; normal distribution; speaker recognition; Baum-Welch algorithm; Gibbs distribution; fuzzy codebooks; fuzzy evolutionary programming; global maxima; hidden Markov model re-estimation; hidden Markov model training; individual speakers; speaker identification; Australia; Covariance matrix; Distributed computing; Genetic programming; Hidden Markov models; Humans; Iterative algorithms; Laboratories; Probability; Speech recognition;
Conference_Titel :
Evolutionary Computation, 1999. CEC 99. Proceedings of the 1999 Congress on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5536-9
DOI :
10.1109/CEC.1999.782505
