Speaker recognition using discrete wavelet transformation

Other Title(s)

تمييز المتحدث باستخدام محول الموجات المنفصل

Dissertant

Abu al-Adas, Firas Id Dayf Allah

Thesis advisor

al-Samawi, Venus Wazir

Comitee Members

Shatnawi, Umar Ali
al-Taani, Ahmad T.
al-Khalidi, Jihad O.

University

Al albayt University

Faculty

Prince Hussein Bin Abdullah Faculty for Information Technology

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2010

English Abstract

The use of biometric information has been known widely for both person identification and security application, it is common knowledge that each person can be identified by the unique characteristics of one or more of person biometrics.

One of the biometric characteristics of that a person can be identified by his voice.

This research attempts to find the best discrete coefficient that can be selected from one or more level using neural network as classifier. The proposed speaker recognition system consists of three phases: preprocessing phase (two processes are performed on the sound, DC level removal and resize of sample for 2000 sample for each sound), feature extraction phase (features that distinguish each sound from another, in this research discrete wavelet transform is used for extracting features from sounds), and recognition phase (many classifiers could be used for speaker recognition, in this research supervised neural networks, Back propagation and LVQ are used as classifiers. This research is concerned with studying the feature extracted from discrete wavelet transformation using feed-forward Back propagation and learning vector quantization as classifiers, the neural networks trained with feature extracted from different levels of discrete wavelet transform (one level at a time).

The research illustrates the effected of using different percent (50 %, 40 %, 30 %, 20 %, 10 %) of each level instead of all feature, then compare their recognition ability and decide the best level in addition to the percent of features that are enough to give comparable results with respect to all feature set and decide the neural network architecture that gives the best recognition rate.

The data set consists of different sounds recorded from ten different persons (three female and seven males) using Microsoft sound recorder, each one pronounce five different statements.

In total, the data set consists of 50 samples (15 female and 35 male).

It was found that 50 % is enough for the recognition process, and level 3, 5 are the best levels to extract features at which the recognition rate reaches 96 % with Back propagation NN.

The system was trained and tested using cross-validation.

Main Subjects

Electronic engineering

Topics

• Algorithms

No. of Pages

50

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-310853