A Method of Wavelet Selection in Phone Recognition

HUGO L. RUFINER; JOHN C. GODDARD

Sacramento, California

Simposio; 40th Midwest Symposium on Circuits and Systems; 1997

IEEE

 In this paper a simple 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.