Repository logo
 

Comparison of effects on subjective intelligibility and quality of speech in babble for two algorithms: A deep recurrent neural network and spectral subtraction.

cam.issuedOnline2019-03-25
dc.contributor.authorKeshavarzi, Mahmoud
dc.contributor.authorGoehring, Tobias
dc.contributor.authorTurner, Richard E
dc.contributor.authorMoore, Brian CJ
dc.contributor.orcidGoehring, Tobias [0000-0002-9038-3310]
dc.contributor.orcidMoore, Brian [0000-0001-7071-0671]
dc.date.accessioned2020-04-16T23:30:28Z
dc.date.available2020-04-16T23:30:28Z
dc.date.issued2019-03
dc.description.abstractThe effects on speech intelligibility and sound quality of two noise-reduction algorithms were compared: a deep recurrent neural network (RNN) and spectral subtraction (SS). The RNN was trained using sentences spoken by a large number of talkers with a variety of accents, presented in babble. Different talkers were used for testing. Participants with mild-to-moderate hearing loss were tested. Stimuli were given frequency-dependent linear amplification to compensate for the individual hearing losses. A paired-comparison procedure was used to compare all possible combinations of three conditions. The conditions were: speech in babble with no processing (NP) or processed using the RNN or SS. In each trial, the same sentence was played twice using two different conditions. The participants indicated which one was better and by how much in terms of speech intelligibility and (in separate blocks) sound quality. Processing using the RNN was significantly preferred over NP and over SS processing for both subjective intelligibility and sound quality, although the magnitude of the preferences was small. SS processing was not significantly preferred over NP for either subjective intelligibility or sound quality. Objective computational measures of speech intelligibility predicted better intelligibility for RNN than for SS or NP.
dc.format.mediumPrint
dc.identifier.doi10.17863/CAM.51516
dc.identifier.eissn1520-8524
dc.identifier.issn0001-4966
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/304436
dc.languageeng
dc.language.isoeng
dc.publisherAcoustical Society of America (ASA)
dc.publisher.urlhttp://dx.doi.org/10.1121/1.5094765
dc.rightsAll rights reserved
dc.subjectAged
dc.subjectFemale
dc.subjectHearing Aids
dc.subjectHumans
dc.subjectMale
dc.subjectMiddle Aged
dc.subjectNeural Networks, Computer
dc.subjectSpeech Intelligibility
dc.subjectSpeech Perception
dc.subjectSpeech Recognition Software
dc.titleComparison of effects on subjective intelligibility and quality of speech in babble for two algorithms: A deep recurrent neural network and spectral subtraction.
dc.typeArticle
dcterms.dateAccepted2019-03-01
prism.issueIdentifier3
prism.publicationDate2019
prism.publicationNameJ Acoust Soc Am
prism.startingPage1493
prism.volume145
pubs.funder-project-idEngineering and Physical Sciences Research Council (EP/L000776/1)
pubs.funder-project-idEngineering and Physical Sciences Research Council (EP/M026957/1)
rioxxterms.licenseref.startdate2019-03
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserved
rioxxterms.typeJournal Article/Review
rioxxterms.versionAM
rioxxterms.versionofrecord10.1121/1.5094765

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
JASA.doc
Size:
366.5 KB
Format:
Microsoft Word
Description:
Accepted version
Licence
http://www.rioxx.net/licenses/all-rights-reserved
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
DepositLicenceAgreementv2.1.pdf
Size:
150.9 KB
Format:
Adobe Portable Document Format