Tree-based Gaussian mixture models for speaker verification

Simple item page
dc.contributor.advisorDu Preez, J. A.
dc.contributor.authorCilliers, Francois Dirken_ZA
dc.contributor.otherUniversity of Stellenbosch. Faculty of Engineering. Dept. of Electrical and Electronic Engineering.
dc.date.accessioned2006-11-01T08:25:54Zen_ZA
dc.date.accessioned2010-06-01T08:29:18Z
dc.date.available2006-11-01T08:25:54Zen_ZA
dc.date.available2010-06-01T08:29:18Z
dc.date.issued2005-12en_ZA
dc.descriptionThesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2005.
dc.description.abstractThe Gaussian mixture model (GMM) performs very effectively in applications such as speech and speaker recognition. However, evaluation speed is greatly reduced when the GMM has a large number of mixture components. Various techniques improve the evaluation speed by reducing the number of required Gaussian evaluations.en_ZA
dc.format.extent1024437 bytesen_ZA
dc.format.mimetypeapplication/pdfen_ZA
dc.identifier.urihttp://hdl.handle.net/10019.1/1639
dc.language.isoen
dc.publisherStellenbosch : University of Stellenbosch
dc.rights.holderUniversity of Stellenbosch
dc.subjectDissertations -- Electronic engineeringen
dc.subjectTheses -- Electronic engineeringen
dc.subjectAutomatic speech recognitionen
dc.subjectSpeech processing systemsen
dc.subject.otherElectrical and Electronic Engineeringen_ZA
dc.titleTree-based Gaussian mixture models for speaker verificationen_ZA
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
cilliers_treebased_2005.pdf
Size:
1000.43 KB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.72 KB
Format:
Plain Text
Description:
Download
Collections
Masters Degrees (Electrical and Electronic Engineering)