Noise-robust speech recognition using top-down selective attention with an HMM classifier

Cited 13 time in webofscience Cited 16 time in scopus
  • Hit : 687
  • Download : 730
DC FieldValueLanguage
dc.contributor.authorLee, CHko
dc.contributor.authorLee, Soo-Youngko
dc.date.accessioned2009-07-23T02:42:05Z-
dc.date.available2009-07-23T02:42:05Z-
dc.date.created2012-02-06-
dc.date.created2012-02-06-
dc.date.issued2007-07-
dc.identifier.citationIEEE SIGNAL PROCESSING LETTERS, v.14, pp.489 - 491-
dc.identifier.issn1070-9908-
dc.identifier.urihttp://hdl.handle.net/10203/10215-
dc.description.abstractFor noise-robust speech recognition, we incorporated a top-down attention mechanism into a hidden Markov model classifier with Mel-frequency cepstral coefficient features. The attention filter was introduced at the outputs of the Mel-scale filterbank and adjusted to maximize the log-likelihood of the attended features with the attended class. A low-complexity constraint was proposed to prevent the attention filter from over-fitting' and a confidence measure was introduced on the attention. A classification was made to the class with the maximum confidence measure, and demonstrated 54 % and 68 % reduction of the false recognition rate with 15-and 20-dB signal-to-noise ratio, respectively.-
dc.languageEnglish-
dc.language.isoen_USen
dc.publisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC-
dc.subjectMODEL-
dc.titleNoise-robust speech recognition using top-down selective attention with an HMM classifier-
dc.typeArticle-
dc.identifier.wosid000247407000015-
dc.identifier.scopusid2-s2.0-34347375731-
dc.type.rimsART-
dc.citation.volume14-
dc.citation.beginningpage489-
dc.citation.endingpage491-
dc.citation.publicationnameIEEE SIGNAL PROCESSING LETTERS-
dc.identifier.doi10.1109/LSP.2006.891326-
dc.contributor.localauthorLee, Soo-Young-
dc.contributor.nonIdAuthorLee, CH-
dc.type.journalArticleArticle-
dc.subject.keywordAuthorhidden Markov model (HMM)-
dc.subject.keywordAuthorselective attention-
dc.subject.keywordAuthorspeech recognition-
dc.subject.keywordPlusMODEL-
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 13 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0