Back to search

University of Waterloo

A Study of the Automatic Speech Recognition Process and Speaker Adaptation

Abstract

dc:description.abstract

This thesis considers the entire automated speech recognition process and presents a standardised approach to LVCSR experimentation with HMMs. It also discusses various approaches to speaker adaptation such as MLLR and multiscale, and presents experimental results for cross­-task speaker adaptation. An analysis of training parameters and data sufficiency for reasonable system performance estimates are also included. It is found that Maximum Likelihood Linear Regression (MLLR) supervised adaptation can result in 6% reduction (absolute) in word error rate given only one minute of adaptation data, as compared with an unadapted model set trained on a different task. The unadapted system performed at 24% WER and the adapted system at 18% WER. This is achieved with only 4 to 7 adaptation classes per speaker, as generated from a regression tree.

Degree

thesis:*
Grantor dc:publisher
University of Waterloo
Year dc:date.issued
2000

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Stokes-Rees, Ian James

Subjects

dc:subject × 7

Rights

dc:rights
Statement dc:rights
  • Copyright: 2000, Stokes-Rees, Ian James. All rights reserved.
Language dc:language.iso
en

Identifiers

dc:identifier.*
Handle dc:identifier.uri
http://hdl.handle.net/10012/840

Chain of custody

source
Harvested from
University of Waterloo
Base URL
dspacemainprd01.lib.uwaterloo.ca/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Stokes-Rees, Ian James. A Study of the Automatic Speech Recognition Process and Speaker Adaptation. University of Waterloo, 2000. http://hdl.handle.net/10012/840