File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Use of periodicity and jitter as speech recognition features

TitleUse of periodicity and jitter as speech recognition features
Authors
Issue Date1998
Citation
Icassp, Ieee International Conference On Acoustics, Speech And Signal Processing - Proceedings, 1998, v. 1, p. 21-24 How to Cite?
AbstractWe investigate a class of features related to voicing parameters that indicate whether the vocal chords are vibrating. Features describing voicing characteristics of speech signals are integrated with an existing 38-dimensional feature vector consisting of first and second order time derivatives of the frame energy and of the cepstral coefficients with their first and second derivatives. HMM-based connected digit recognition experiments comparing the traditional and extended feature sets show that voicing features and spectral information are complementary and that improved speech recognition performance is obtained by combining the two sources of information.
Persistent Identifierhttp://hdl.handle.net/10722/179577
ISSN

 

DC FieldValueLanguage
dc.contributor.authorThomson, David Len_US
dc.contributor.authorChengalvarayan, Rathinaveluen_US
dc.date.accessioned2012-12-19T09:59:57Z-
dc.date.available2012-12-19T09:59:57Z-
dc.date.issued1998en_US
dc.identifier.citationIcassp, Ieee International Conference On Acoustics, Speech And Signal Processing - Proceedings, 1998, v. 1, p. 21-24en_US
dc.identifier.issn0736-7791en_US
dc.identifier.urihttp://hdl.handle.net/10722/179577-
dc.description.abstractWe investigate a class of features related to voicing parameters that indicate whether the vocal chords are vibrating. Features describing voicing characteristics of speech signals are integrated with an existing 38-dimensional feature vector consisting of first and second order time derivatives of the frame energy and of the cepstral coefficients with their first and second derivatives. HMM-based connected digit recognition experiments comparing the traditional and extended feature sets show that voicing features and spectral information are complementary and that improved speech recognition performance is obtained by combining the two sources of information.en_US
dc.languageengen_US
dc.relation.ispartofICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedingsen_US
dc.titleUse of periodicity and jitter as speech recognition featuresen_US
dc.typeConference_Paperen_US
dc.identifier.emailThomson, David L: dthomson@hku.hken_US
dc.identifier.authorityThomson, David L=rp00788en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.scopuseid_2-s2.0-0031643033en_US
dc.identifier.volume1en_US
dc.identifier.spage21en_US
dc.identifier.epage24en_US
dc.identifier.scopusauthoridThomson, David L=7202586830en_US
dc.identifier.scopusauthoridChengalvarayan, Rathinavelu=6701843465en_US

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats