Cover Image

Feature extraction and classification of malay speech vowels

In human language, a phoneme is the smallest structural unit that distinguishes meaning. Normally, language like English commonly combines phonemes to form a word. In many languages, the Consonant-Vowel (CV) units have the highest frequency of occurrence among different forms of subword units. Th...

Full description

Saved in:
Bibliographic Details
Main Author: Shahrul Azmi, Mohd Yusof
Format: Thesis
Language:English
Published: Universiti Malaysia Perlis 2011
Subjects:
Consonant-Vowel (CV)
Vowel recognition
Automatic Speech Recognition (ASR)
Linear Discriminat Analysis (LDA)
Robustness analysis
Malays vowel analysis
Online Access:http://dspace.unimap.edu.my/xmlui/handle/123456789/12916
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.unimap-12916
record_format dspace
spelling my.unimap-129162011-07-01T04:13:24Z Feature extraction and classification of malay speech vowels Shahrul Azmi, Mohd Yusof Consonant-Vowel (CV) Vowel recognition Automatic Speech Recognition (ASR) Linear Discriminat Analysis (LDA) Robustness analysis Malays vowel analysis In human language, a phoneme is the smallest structural unit that distinguishes meaning. Normally, language like English commonly combines phonemes to form a word. In many languages, the Consonant-Vowel (CV) units have the highest frequency of occurrence among different forms of subword units. Therefore, recognition of CV units with a good accuracy is crucial for development of a speech recognition system. There are also many applications that use vowels phonemes. Among them are speech therapy systems that improve utterances of word pronunciation especially to children. There are also systems that teach hearing impaired person to speak properly by pronouncing words with a good degree of intelligibility. All of these systems require high degree of vowel recognition capability in which this study focuses on. This thesis contributes five modified feature extraction methods for vowel recognition based on intensities of the Frequency Filter Bands. They are First Formant Bandwidth (F1BW), Fixed Formant Frequency Band (FFB), Spectral Delta (SpD), Bark Intensity (BrKI) and Formant Frequency Difference (FFD). The performance of these five proposed methods are compared with performance of three conventional feature extraction methods of single frame Mel-frequency cepstrum coefficients (MFCCs), multiple frame Mel-frequency cepstrum coefficients (MFCCf) and the first three formant features. The classifiers analysed in this study were Multinomial Logistic Regression (MLR), Levenberg-Marquardt (LM) network, k-Nearest Neighbors (KNN) and Linear Discriminant Analysis (LDA). There are four main contributions of this thesis. First is the new vowel corpus consisting of more than 1300 recorded vowels from 100 Malaysian speakers. Second are the five improved feature extraction methods which perform better than MFCC on single frame analysis. The third is the performance and robustness analysis using different classifiers and different Gaussian noise level. The fourth contribution is the frame analysis criteria for isolated vowel analysis. 2011-07-01T04:13:24Z 2011-07-01T04:13:24Z 2010 Thesis http://hdl.handle.net/123456789/12916 en Universiti Malaysia Perlis School of Mechatronic Engineering
institution Universiti Malaysia Perlis
building UniMAP Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Malaysia Perlis
content_source UniMAP Library Digital Repository
url_provider http://dspace.unimap.edu.my/
language English
topic Consonant-Vowel (CV)
Vowel recognition
Automatic Speech Recognition (ASR)
Linear Discriminat Analysis (LDA)
Robustness analysis
Malays vowel analysis
spellingShingle Consonant-Vowel (CV)
Vowel recognition
Automatic Speech Recognition (ASR)
Linear Discriminat Analysis (LDA)
Robustness analysis
Malays vowel analysis
Shahrul Azmi, Mohd Yusof
Feature extraction and classification of malay speech vowels
description In human language, a phoneme is the smallest structural unit that distinguishes meaning. Normally, language like English commonly combines phonemes to form a word. In many languages, the Consonant-Vowel (CV) units have the highest frequency of occurrence among different forms of subword units. Therefore, recognition of CV units with a good accuracy is crucial for development of a speech recognition system. There are also many applications that use vowels phonemes. Among them are speech therapy systems that improve utterances of word pronunciation especially to children. There are also systems that teach hearing impaired person to speak properly by pronouncing words with a good degree of intelligibility. All of these systems require high degree of vowel recognition capability in which this study focuses on. This thesis contributes five modified feature extraction methods for vowel recognition based on intensities of the Frequency Filter Bands. They are First Formant Bandwidth (F1BW), Fixed Formant Frequency Band (FFB), Spectral Delta (SpD), Bark Intensity (BrKI) and Formant Frequency Difference (FFD). The performance of these five proposed methods are compared with performance of three conventional feature extraction methods of single frame Mel-frequency cepstrum coefficients (MFCCs), multiple frame Mel-frequency cepstrum coefficients (MFCCf) and the first three formant features. The classifiers analysed in this study were Multinomial Logistic Regression (MLR), Levenberg-Marquardt (LM) network, k-Nearest Neighbors (KNN) and Linear Discriminant Analysis (LDA). There are four main contributions of this thesis. First is the new vowel corpus consisting of more than 1300 recorded vowels from 100 Malaysian speakers. Second are the five improved feature extraction methods which perform better than MFCC on single frame analysis. The third is the performance and robustness analysis using different classifiers and different Gaussian noise level. The fourth contribution is the frame analysis criteria for isolated vowel analysis.
format Thesis
author Shahrul Azmi, Mohd Yusof
author_facet Shahrul Azmi, Mohd Yusof
author_sort Shahrul Azmi, Mohd Yusof
title Feature extraction and classification of malay speech vowels
title_short Feature extraction and classification of malay speech vowels
title_full Feature extraction and classification of malay speech vowels
title_fullStr Feature extraction and classification of malay speech vowels
title_full_unstemmed Feature extraction and classification of malay speech vowels
title_sort feature extraction and classification of malay speech vowels
publisher Universiti Malaysia Perlis
publishDate 2011
url http://dspace.unimap.edu.my/xmlui/handle/123456789/12916
_version_ 1643790698448683008
score 13.222552

Similar Items