File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A Study of Minimum Classification Error (MCE) Linear Regression for Supervised Adaptation of MCE-Trained Continuous-Density Hidden Markov Models

TitleA Study of Minimum Classification Error (MCE) Linear Regression for Supervised Adaptation of MCE-Trained Continuous-Density Hidden Markov Models
Authors
Wu, JHuo, Q
KeywordsHidden Markov model (HMM)
HMM adaptation
minimum classification error linear regression (MCELR)
speaker adaptation
Issue Date2007
PublisherIEEE.
Citation
IEEE Transactions on Audio, Speech and Language Processing, 2007, v. 15 n. 2, p. 478-488 How to Cite?
DOI: http://dx.doi.org/10.1109/TASL.2006.881692
AbstractIn this paper, we present a formulation of minimum classification error linear regression (MCELR) for the adaptation of Gaussian mixture continuous-density hidden Markov model (CDHMM) parameters. Two optimization approaches, namely generalized probabilistic descent (GPD) and Quickprop are studied and compared for the optimization of the MCELR objective function. The effectiveness of the proposed MCELR technique is confirmed via a series of supervised speaker adaptation experiments on a task of continuous Putonghua (Mandarin Chinese) speech recognition.
Persistent Identifierhttp://hdl.handle.net/10722/47082
ISSN
1558-7916
2015 Impact Factor: 1.877
ISI Accession Number ID
WOS:000243914800010

 

DC FieldValueLanguage
dc.contributor.authorWu, Jen_HK
dc.contributor.authorHuo, Qen_HK
dc.date.accessioned2007-10-30T07:06:42Z-
dc.date.available2007-10-30T07:06:42Z-
dc.date.issued2007en_HK
dc.identifier.citationIEEE Transactions on Audio, Speech and Language Processing, 2007, v. 15 n. 2, p. 478-488en_HK
dc.identifier.issn1558-7916en_HK
dc.identifier.urihttp://hdl.handle.net/10722/47082-
dc.description.abstractIn this paper, we present a formulation of minimum classification error linear regression (MCELR) for the adaptation of Gaussian mixture continuous-density hidden Markov model (CDHMM) parameters. Two optimization approaches, namely generalized probabilistic descent (GPD) and Quickprop are studied and compared for the optimization of the MCELR objective function. The effectiveness of the proposed MCELR technique is confirmed via a series of supervised speaker adaptation experiments on a task of continuous Putonghua (Mandarin Chinese) speech recognition.en_HK
dc.format.extent365979 bytes-
dc.format.extent1860 bytes-
dc.format.extent7254 bytes-
dc.format.mimetypeapplication/pdf-
dc.format.mimetypetext/plain-
dc.format.mimetypetext/plain-
dc.languageengen_HK
dc.publisherIEEE.en_HK
dc.relation.ispartofIEEE Transactions on Audio, Speech, and Language Processing-
dc.rights©2007 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.-
dc.subjectHidden Markov model (HMM)en_HK
dc.subjectHMM adaptationen_HK
dc.subjectminimum classification error linear regression (MCELR)en_HK
dc.subjectspeaker adaptationen_HK
dc.titleA Study of Minimum Classification Error (MCE) Linear Regression for Supervised Adaptation of MCE-Trained Continuous-Density Hidden Markov Modelsen_HK
dc.typeArticleen_HK
dc.identifier.openurlhttp://library.hku.hk:4550/resserv?sid=HKU:IR&issn=1558-7916&volume=15&issue=2&spage=478&epage=488&date=2007&atitle=A+Study+of+Minimum+Classification+Error+(MCE)+Linear+Regression+for+Supervised+Adaptation+of+MCE-Trained+Continuous-Density+Hidden+Markov+Modelsen_HK
dc.description.naturepublished_or_final_versionen_HK
dc.identifier.doi10.1109/TASL.2006.881692en_HK
dc.identifier.scopuseid_2-s2.0-58349123022-
dc.identifier.hkuros126453-
dc.identifier.isiWOS:000243914800010-
dc.identifier.issnl1558-7916-

Export via OAI-PMH Interface in XML Formats

OR

Export to Other Non-XML Formats