Tree-based Gaussian mixture models for speaker verification

Tree-based Gaussian mixture models for speaker verification

Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2005. === 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...

Full description

Bibliographic Details
Main Author: Cilliers, Francois Dirk
Other Authors: Du Preez, J. A.
Format: Others
Language:en
Published: Stellenbosch : University of Stellenbosch 2006
Subjects:
Dissertations > Electronic engineering
Theses > Electronic engineering
Automatic speech recognition
Speech processing systems
Electrical and Electronic Engineering
Online Access:http://hdl.handle.net/10019.1/1639
Description
Summary:Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2005. === 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.

Similar Items