Channel-weighting method for speech recognition using wavelet decompositions

Jyh Shing Shyuu, Jhing Fa Wang, Chung-Hsien Wu

Research output: Contribution to conferencePaper

2 Citations (Scopus)

Abstract

A decomposition of signal into a set of frequency channels of equal bandwidth on a logarithmic scale, i.e., an analysis of the signal using constant Q filters, using wavelet and multiresolution analysis is used in this paper to derive cepstrum features of different spatial frequency bands. Based on the decompositions, each channel is modeled as a Bayesian subnetwork and each subnetwork is weighted by a weighting algorithm. The distortions for speech recognition between a reference model and the input vectors are then computed by summing the weighted scores of all decomposed channels. The experimental results show that the recognition rate of this method is superior to those non-weighting methods.

Original languageEnglish
Pages519-523
Number of pages5
Publication statusPublished - 1994 Dec 1
EventProceedings of the 1994 IEEE Asia-Pacific Conference on Circuits and Systems - Taipei, Taiwan
Duration: 1994 Dec 51994 Dec 8

Other

OtherProceedings of the 1994 IEEE Asia-Pacific Conference on Circuits and Systems
CityTaipei, Taiwan
Period94-12-0594-12-08

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Channel-weighting method for speech recognition using wavelet decompositions'. Together they form a unique fingerprint.

Cite this