A method of wavelet selection in phoneme recognition

Author

Rufiner, Hugo L. ; C, John Goddard

Author_Institution

Lab. de Cibernetica, Univ. Nacional de Entre Rios, Argentina

Volume

2

fYear

1997

fDate

3-6 Aug 1997

Firstpage

889

Abstract

In this paper a method is proposed for choosing between different wavelets, and their corresponding parameters, and it is applied to TIMIT phoneme classification. The method involves the use of a Kohonen network to extract prototypes for each class. Distance measures between them are used as criteria for choosing the best wavelet. In the case of phoneme classification, a time delay neural network is used as the classifier. A high correlation is found between results of the proposed method and the classification accuracy obtained by the network

Keywords

pattern classification; self-organising feature maps; speech recognition; wavelet transforms; Kohonen network; TIMIT phoneme classification; classification accuracy; distance measures; phoneme recognition; time delay neural network; wavelet selection; Approximation algorithms; Fourier transforms; Matching pursuit algorithms; Pursuit algorithms; Signal analysis; Signal processing algorithms; Speech analysis; Wavelet analysis; Wavelet packets; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1997. Proceedings of the 40th Midwest Symposium on

Conference_Location

Sacramento, CA

Print_ISBN

0-7803-3694-1

Type

conf

DOI

10.1109/MWSCAS.1997.662217

Filename

662217