Show simple item record

dc.contributor.authorStark, Anthonyen_US
dc.contributor.authorPaliwal, Kuldipen_US
dc.date.accessioned2017-05-03T13:12:00Z
dc.date.available2017-05-03T13:12:00Z
dc.date.issued2011en_US
dc.date.modified2012-02-14T05:33:49Z
dc.identifier.issn01676393en_US
dc.identifier.doi10.1016/j.specom.2010.08.001en_US
dc.identifier.urihttp://hdl.handle.net/10072/42592
dc.description.abstractIn this paper, we investigate the use of the minimum mean square error (MMSE) spectral energy estimator for use in environmentrobust automatic speech recognition (ASR). In the past, it has been common to use the MMSE log-spectral amplitude estimator for this task. However, this estimator was originally derived under subjective human listening criteria. Therefore its complex suppression rule may not be optimal for use in ASR. On the other hand, it can be shown that the MMSE spectral energy estimator is closely related to the MMSE Mel-frequency cepstral coefficient (MFCC) estimator. Despite this, the spectral energy estimator has tended to suffer from the problem of excessive residual noise. We examine the cause of this residual noise and show that the introduction of a heuristic based speech presence uncertainty (SPU) can significantly improve its performance as a front-end ASR enhancement regime. The proposed spectral energy SPU estimator is evaluated on the Aurora2, RM and OLLO2 speech recognition tasks and can be shown to significantly improve additive noise robustness over the more common spectral amplitude and log-spectral amplitude estimators.en_US
dc.description.peerreviewedYesen_US
dc.description.publicationstatusYesen_US
dc.languageEnglishen_US
dc.publisherElsevieren_US
dc.publisher.placeNetherlandsen_US
dc.relation.ispartofstudentpublicationNen_US
dc.relation.ispartofpagefrom51en_US
dc.relation.ispartofpageto61en_US
dc.relation.ispartofjournalSpeech Communicationen_US
dc.relation.ispartofvolume53en_US
dc.rights.retentionYen_US
dc.subject.fieldofresearchSignal Processingen_US
dc.subject.fieldofresearchcode090609en_US
dc.titleUse of speech presence uncertainty with MMSE spectral energy estimation for robust automatic speech recognitionen_US
dc.typeJournal articleen_US
dc.type.descriptionC1 - Peer Reviewed (HERDC)en_US
dc.type.codeC - Journal Articlesen_US
gro.facultyGriffith Sciences, Griffith School of Engineeringen_US
gro.date.issued2011
gro.hasfulltextNo Full Text


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

  • Journal articles
    Contains articles published by Griffith authors in scholarly journals.

Show simple item record