File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Bayesian adaptive learning of the parameters of hidden Markov model for speech recognition

TitleBayesian adaptive learning of the parameters of hidden Markov model for speech recognition
Authors
Issue Date1995
PublisherIEEE.
Citation
IEEE Transactions on Speech and Audio Processing, 1995, v. 3 n. 5, p. 334-345 How to Cite?
AbstractA theoretical framework for Bayesian adaptive training of the parameters of a discrete hidden Markov model (DHMM) and of a semi-continuous HMM (SCHMM) with Gaussian mixture state observation densities is presented. In addition to formulating the forward-backward MAP (maximum a posteriori) and the segmental MAP algorithms for estimating the above HMM parameters, a computationally efficient segmental quasi-Bayes algorithm for estimating the state-specific mixture coefficients in SCHMM is developed. For estimating the parameters of the prior densities, a new empirical Bayes method based on the moment estimates is also proposed. The MAP algorithms and the prior parameter specification are directly applicable to training speaker adaptive HMMs. Practical issues related to the use of the proposed techniques for HMM-based speaker adaptation are studied. The proposed MAP algorithms are shown to be effective especially in the cases in which the training or adaptation data are limited.
Persistent Identifierhttp://hdl.handle.net/10722/43667
ISSN
2007 Impact Factor: 2.291

 

DC FieldValueLanguage
dc.contributor.authorHuo, Qen_HK
dc.contributor.authorChan, Cen_HK
dc.contributor.authorLee, CHen_HK
dc.date.accessioned2007-03-23T04:51:36Z-
dc.date.available2007-03-23T04:51:36Z-
dc.date.issued1995en_HK
dc.identifier.citationIEEE Transactions on Speech and Audio Processing, 1995, v. 3 n. 5, p. 334-345en_HK
dc.identifier.issn1063-6676en_HK
dc.identifier.urihttp://hdl.handle.net/10722/43667-
dc.description.abstractA theoretical framework for Bayesian adaptive training of the parameters of a discrete hidden Markov model (DHMM) and of a semi-continuous HMM (SCHMM) with Gaussian mixture state observation densities is presented. In addition to formulating the forward-backward MAP (maximum a posteriori) and the segmental MAP algorithms for estimating the above HMM parameters, a computationally efficient segmental quasi-Bayes algorithm for estimating the state-specific mixture coefficients in SCHMM is developed. For estimating the parameters of the prior densities, a new empirical Bayes method based on the moment estimates is also proposed. The MAP algorithms and the prior parameter specification are directly applicable to training speaker adaptive HMMs. Practical issues related to the use of the proposed techniques for HMM-based speaker adaptation are studied. The proposed MAP algorithms are shown to be effective especially in the cases in which the training or adaptation data are limited.en_HK
dc.format.extent1283284 bytes-
dc.format.extent2668 bytes-
dc.format.mimetypeapplication/pdf-
dc.format.mimetypetext/plain-
dc.languageengen_HK
dc.publisherIEEE.en_HK
dc.rightsCreative Commons: Attribution 3.0 Hong Kong License-
dc.rights©1995 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.en_HK
dc.titleBayesian adaptive learning of the parameters of hidden Markov model for speech recognitionen_HK
dc.typeArticleen_HK
dc.identifier.openurlhttp://library.hku.hk:4550/resserv?sid=HKU:IR&issn=1063-6676&volume=3&issue=5&spage=334&epage=345&date=1995&atitle=Bayesian+adaptive+learning+of+the+parameters+of+hidden+Markov+model+for+speech+recognitionen_HK
dc.description.naturepublished_or_final_versionen_HK
dc.identifier.doi10.1109/89.466661-
dc.identifier.scopuseid_2-s2.0-0029377113-
dc.identifier.hkuros8368-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats