Comparison of speech recognition techniques in a noisy environment

In most speech recognition evaluation studies, the focus is usually on one single speech recognition technique such as Discrete Hidden Markov Modeling with codebooks of different sizes. Inter-technique comparison study using a com-mon database has not been systematically explored. In this thesis, a...

Full description

Saved in:
Bibliographic Details
Main Author: Tan, Chin Tuan.
Other Authors: Chang, Joseph Sylvester
Format: Theses and Dissertations
Language:English
Published: 2009
Subjects:
Online Access:http://hdl.handle.net/10356/19688
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:In most speech recognition evaluation studies, the focus is usually on one single speech recognition technique such as Discrete Hidden Markov Modeling with codebooks of different sizes. Inter-technique comparison study using a com-mon database has not been systematically explored. In this thesis, a com-parative study on three speech recognition techniques namely, Dynamic Time Warping (DTW), Discrete Hidden Markov Modeling (DHMM) and Neural Net-work(Multilayer Perceptron) (NN(MLP)) was conducted using one common database.