Tree-based Gaussian mixture models for speaker verification
dc.contributor.advisor | Du Preez, J. A. | |
dc.contributor.author | Cilliers, Francois Dirk | en_ZA |
dc.contributor.other | University of Stellenbosch. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. | |
dc.date.accessioned | 2006-11-01T08:25:54Z | en_ZA |
dc.date.accessioned | 2010-06-01T08:29:18Z | |
dc.date.available | 2006-11-01T08:25:54Z | en_ZA |
dc.date.available | 2010-06-01T08:29:18Z | |
dc.date.issued | 2005-12 | en_ZA |
dc.description | Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2005. | |
dc.description.abstract | The 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.extent | 1024437 bytes | en_ZA |
dc.format.mimetype | application/pdf | en_ZA |
dc.identifier.uri | http://hdl.handle.net/10019.1/1639 | |
dc.language.iso | en | |
dc.publisher | Stellenbosch : University of Stellenbosch | |
dc.rights.holder | University of Stellenbosch | |
dc.subject | Dissertations -- Electronic engineering | en |
dc.subject | Theses -- Electronic engineering | en |
dc.subject | Automatic speech recognition | en |
dc.subject | Speech processing systems | en |
dc.subject.other | Electrical and Electronic Engineering | en_ZA |
dc.title | Tree-based Gaussian mixture models for speaker verification | en_ZA |
dc.type | Thesis | en_ZA |