Speech recognition enhancement using beamforming and a genetic algorithm

K. Y. Chan; S. Y. Low; S. Nordholm; Ka Fai Cedric Yiu; S. H. Ling

doi:10.1109/NSS.2009.44

Speech recognition enhancement using beamforming and a genetic algorithm

K. Y. Chan, S. Y. Low, S. Nordholm, Ka Fai Cedric Yiu, S. H. Ling

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

5 Citations (Scopus)

Abstract

This paper proposes a genetic algorithm (GA) based beamformer to optimize speech recognition accuracy for a pre-trained speech recognizer. The proposed beamformer is designed to tackle the non-differentiable and non-linear natures of speech recognition by employing the GA algorithm to search for the optimal beamformer weights. Specifically, a population of beamformer weights is reproduced by crossover and mutation until the optimal beamformer weights are obtained. Results show that the speech recognition accuracies can be greatly improved even in noisy environments.

Original language	English
Title of host publication	NSS 2009 - Network and System Security
Pages	510-515
Number of pages	6
DOIs	https://doi.org/10.1109/NSS.2009.44
Publication status	Published - 1 Dec 2009
Event	2009 3rd International Conference on Network and System Security, NSS 2009 - Gold Coast, QLD, Australia Duration: 19 Oct 2009 → 21 Oct 2009

Conference

Conference	2009 3rd International Conference on Network and System Security, NSS 2009
Country/Territory	Australia
City	Gold Coast, QLD
Period	19/10/09 → 21/10/09

Keywords

Beamforming
Genetic algorithm
Signal enhancement
Speech recognition

ASJC Scopus subject areas

Computer Networks and Communications
Software
Control and Systems Engineering

More information

10.1109/NSS.2009.44

Cite this

@inproceedings{f54abd21c4924772aa714eb7d59edbac,

title = "Speech recognition enhancement using beamforming and a genetic algorithm",

abstract = "This paper proposes a genetic algorithm (GA) based beamformer to optimize speech recognition accuracy for a pre-trained speech recognizer. The proposed beamformer is designed to tackle the non-differentiable and non-linear natures of speech recognition by employing the GA algorithm to search for the optimal beamformer weights. Specifically, a population of beamformer weights is reproduced by crossover and mutation until the optimal beamformer weights are obtained. Results show that the speech recognition accuracies can be greatly improved even in noisy environments.",

keywords = "Beamforming, Genetic algorithm, Signal enhancement, Speech recognition",

author = "Chan, {K. Y.} and Low, {S. Y.} and S. Nordholm and Yiu, {Ka Fai Cedric} and Ling, {S. H.}",

year = "2009",

month = dec,

day = "1",

doi = "10.1109/NSS.2009.44",

language = "English",

isbn = "9780769538389",

pages = "510--515",

booktitle = "NSS 2009 - Network and System Security",

note = "2009 3rd International Conference on Network and System Security, NSS 2009 ; Conference date: 19-10-2009 Through 21-10-2009",

}

TY - GEN

T1 - Speech recognition enhancement using beamforming and a genetic algorithm

AU - Chan, K. Y.

AU - Low, S. Y.

AU - Nordholm, S.

AU - Yiu, Ka Fai Cedric

AU - Ling, S. H.

PY - 2009/12/1

Y1 - 2009/12/1

N2 - This paper proposes a genetic algorithm (GA) based beamformer to optimize speech recognition accuracy for a pre-trained speech recognizer. The proposed beamformer is designed to tackle the non-differentiable and non-linear natures of speech recognition by employing the GA algorithm to search for the optimal beamformer weights. Specifically, a population of beamformer weights is reproduced by crossover and mutation until the optimal beamformer weights are obtained. Results show that the speech recognition accuracies can be greatly improved even in noisy environments.

AB - This paper proposes a genetic algorithm (GA) based beamformer to optimize speech recognition accuracy for a pre-trained speech recognizer. The proposed beamformer is designed to tackle the non-differentiable and non-linear natures of speech recognition by employing the GA algorithm to search for the optimal beamformer weights. Specifically, a population of beamformer weights is reproduced by crossover and mutation until the optimal beamformer weights are obtained. Results show that the speech recognition accuracies can be greatly improved even in noisy environments.

KW - Beamforming

KW - Genetic algorithm

KW - Signal enhancement

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=72849141934&partnerID=8YFLogxK

U2 - 10.1109/NSS.2009.44

DO - 10.1109/NSS.2009.44

M3 - Conference article published in proceeding or book

SN - 9780769538389

SP - 510

EP - 515

BT - NSS 2009 - Network and System Security

T2 - 2009 3rd International Conference on Network and System Security, NSS 2009

Y2 - 19 October 2009 through 21 October 2009

ER -

Speech recognition enhancement using beamforming and a genetic algorithm

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this