Tree-based Gaussian mixture models for speaker verification
Date
2005-12
Authors
Cilliers, Francois Dirk
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : University of Stellenbosch
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.
Description
Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2005.
Keywords
Dissertations -- Electronic engineering, Theses -- Electronic engineering, Automatic speech recognition, Speech processing systems