File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Speech recognition enhancement using beamforming and a genetic algorithm

TitleSpeech recognition enhancement using beamforming and a genetic algorithm
Authors
Chan, KYLow, SYNordholm, SYiu, KFCLing, SH
KeywordsBeamforming
Genetic Algorithm
Signal Enhancement
Speech Recognition
Issue Date2009
Citation
Nss 2009 - Network And System Security, 2009, p. 510-515 How to Cite?
DOI: http://dx.doi.org/10.1109/NSS.2009.44
AbstractThis 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. © 2009 IEEE.
Persistent Identifierhttp://hdl.handle.net/10722/155924
References
References in Scopus

 

DC FieldValueLanguage
dc.contributor.authorChan, KYen_US
dc.contributor.authorLow, SYen_US
dc.contributor.authorNordholm, Sen_US
dc.contributor.authorYiu, KFCen_US
dc.contributor.authorLing, SHen_US
dc.date.accessioned2012-08-08T08:38:26Z-
dc.date.available2012-08-08T08:38:26Z-
dc.date.issued2009en_US
dc.identifier.citationNss 2009 - Network And System Security, 2009, p. 510-515en_US
dc.identifier.urihttp://hdl.handle.net/10722/155924-
dc.description.abstractThis 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. © 2009 IEEE.en_US
dc.languageengen_US
dc.relation.ispartofNSS 2009 - Network and System Securityen_US
dc.subjectBeamformingen_US
dc.subjectGenetic Algorithmen_US
dc.subjectSignal Enhancementen_US
dc.subjectSpeech Recognitionen_US
dc.titleSpeech recognition enhancement using beamforming and a genetic algorithmen_US
dc.typeArticleen_US
dc.identifier.emailYiu, KFC:cedric@hkucc.hku.hken_US
dc.identifier.authorityYiu, KFC=rp00206en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.doi10.1109/NSS.2009.44en_US
dc.identifier.scopuseid_2-s2.0-72849141934en_US
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-72849141934&selection=ref&src=s&origin=recordpageen_US
dc.identifier.spage510en_US
dc.identifier.epage515en_US
dc.identifier.scopusauthoridChan, KY=25639498200en_US
dc.identifier.scopusauthoridLow, SY=7102636488en_US
dc.identifier.scopusauthoridNordholm, S=7005690573en_US
dc.identifier.scopusauthoridYiu, KFC=24802813000en_US
dc.identifier.scopusauthoridLing, SH=24342010000en_US

Export via OAI-PMH Interface in XML Formats

OR

Export to Other Non-XML Formats