Speaker Recognition Using Multiple Parametric Self-Organizing Maps

Authors

Pablo Gomez, Florida International University

Document Type

Thesis

Major/Program

Electrical Engineering

First Advisor's Name

Kang Yen

First Advisor's Committee Title

Committee Chair

Second Advisor's Name

Jean Andrian

Second Advisor's Committee Title

Committee Member

Third Advisor's Name

Armando Barreto

Third Advisor's Committee Title

Committee Member

Date of Defense

7-28-2003

Abstract

Speaker Recognition is the process of automatically recognizing a person who is speaking on the basis of individual parameters included in his/her voice. This technology allows systems to automatically verify identify in applications such as banking by telephone or forensic science.

A Speaker Recognition system has the following main modules: Feature Extraction and Classification.

For feature extraction the most commonly used techniques are MEL-Frequency Cepstrum Coefficients (MFCC) and Linear Predictive Coding (LPC). For classification and verification, technologies such as Vector Quantization (VQ), Hidden Markov Models (HMM) and Neural Networks have been used.

The contribution of this thesis is a new methodology to achieve high accuracy identification and impostor rejection. The new proposed method, Multiple Parametric Self-Organizing Maps (M-PSOM) is a classification and verification technique. The new method was successfully implemented and tested using the CSLU Speaker Recognition Corpora of the Oregon School of Engineering with excellent results.

Identifier

FI15101622

Recommended Citation

Gomez, Pablo, "Speaker Recognition Using Multiple Parametric Self-Organizing Maps" (2003). FIU Electronic Theses and Dissertations. 4763.
https://digitalcommons.fiu.edu/etd/4763