Channel-weighting method for speech recognition using wavelet decompositions

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

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (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
Title of host publicationIEEE Asia-Pacific Conference on Circuits and Systems - Proceedings
PublisherIEEE
Pages519-523
Number of pages5
Publication statusPublished - 1994
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

Fingerprint

Multiresolution analysis
Wavelet decomposition
Wavelet analysis
Speech recognition
Frequency bands
Bandwidth

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

Shyuu, J. S., Wang, J. F., & Wu, C-H. (1994). Channel-weighting method for speech recognition using wavelet decompositions. In IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings (pp. 519-523). IEEE.
Shyuu, Jyh Shing ; Wang, Jhing Fa ; Wu, Chung-Hsien. / Channel-weighting method for speech recognition using wavelet decompositions. IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings. IEEE, 1994. pp. 519-523
@inproceedings{15a2ccd107514f369c5cb504ac1b2823,
title = "Channel-weighting method for speech recognition using wavelet decompositions",
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.",
author = "Shyuu, {Jyh Shing} and Wang, {Jhing Fa} and Chung-Hsien Wu",
year = "1994",
language = "English",
pages = "519--523",
booktitle = "IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings",
publisher = "IEEE",

}

Shyuu, JS, Wang, JF & Wu, C-H 1994, Channel-weighting method for speech recognition using wavelet decompositions. in IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings. IEEE, pp. 519-523, Proceedings of the 1994 IEEE Asia-Pacific Conference on Circuits and Systems, Taipei, Taiwan, 94-12-05.

Channel-weighting method for speech recognition using wavelet decompositions. / Shyuu, Jyh Shing; Wang, Jhing Fa; Wu, Chung-Hsien.

IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings. IEEE, 1994. p. 519-523.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Channel-weighting method for speech recognition using wavelet decompositions

AU - Shyuu, Jyh Shing

AU - Wang, Jhing Fa

AU - Wu, Chung-Hsien

PY - 1994

Y1 - 1994

N2 - 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.

AB - 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.

UR - http://www.scopus.com/inward/record.url?scp=0028740144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028740144&partnerID=8YFLogxK

M3 - Conference contribution

SP - 519

EP - 523

BT - IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings

PB - IEEE

ER -

Shyuu JS, Wang JF, Wu C-H. Channel-weighting method for speech recognition using wavelet decompositions. In IEEE Asia-Pacific Conference on Circuits and Systems - Proceedings. IEEE. 1994. p. 519-523