Award Date

1-1-2006

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

First Committee Member

Evangelos Yfantis

Number of Pages

49

Abstract

This research mainly focuses on recognizing the speakers through their speech samples. Numerous "Text-Dependent" or "Text-Independent" algorithms have been developed by people so far, to recognize the speaker from his/her speech. In this thesis, we concentrate on the recognition of the speaker from the fixed text i.e. "Text-Dependent". Possibility of extending this method to variable text i.e. "Text-Independent" is also analyzed. Different feature extraction algorithms are employed and their performance with Artificial Neural Networks as a Data Classifier on a fixed training set is analyzed. We find a way to combine all these individual feature extraction algorithms by incorporating their interdependence. The efficiency of these algorithms is determined after the input speech is classified using Back Propagation Algorithm of Artificial Neural Networks. A special case of Back Propagation Algorithm which improves the efficiency of the classification is also discussed.

Keywords

Artificial; Identification; Networks; Neural; Robust; Speaker

Controlled Subject

Computer science; Electrical engineering

File Format

pdf

File Size

1269.76 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Permissions

If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to digitalscholarship@unlv.edu and include clear identification of the work, preferably with URL.

Rights

IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/

Download


COinS