Cover Image

Vowel recognition based on frequency ranges determined by bandwidth approach

Link to publisher's homepage at http://ieeexplore.ieee.org

Saved in:
Bibliographic Details
Main Authors: Paulraj, M. P., Sazali, Yaacob, Mohd Yusof, S. A.
Format: Article
Language:English
Published: Institute of Eelectrical and Electronics Engineering (IEEE) 2009
Subjects:
Feature extraction
Speech recognition
Natural language processing
Multilayer perceptrons
Digital electronics
Signal theory (Telecommunication)
Automatic speech recognition
Speech perception
Online Access:http://dspace.unimap.edu.my/xmlui/handle/123456789/6869
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.unimap-6869
record_format dspace
spelling my.unimap-68692010-11-23T06:35:54Z Vowel recognition based on frequency ranges determined by bandwidth approach Paulraj, M. P. Sazali, Yaacob Mohd Yusof, S. A. Feature extraction Speech recognition Natural language processing Multilayer perceptrons Digital electronics Signal theory (Telecommunication) Automatic speech recognition Speech perception Link to publisher's homepage at http://ieeexplore.ieee.org Automatic speech recognition (ASR) has made great strides with the development of digital signal processing hardware and software especially using English as the language of choice. In this paper, a new feature extraction method is presented to identify vowels recorded from 80 Malaysian speakers. The features were obtained from Vocal Tract Model based on Bandwidth (BW) approach. Bandwidth approach identifies frequency bands based on the first peak of vowel frequency responses. Mean and maximum energies were calculated from these Bandwidth frequency bands. Classification results from Bandwidth Approach were compared with the first 3-formant features using Linear Predictive method. A Multi-Layer Perceptron (MLP) and Multinomial Logistic Regression (MLR) were used to classify the vowels. MLR and MLP shows comparable classification results for BW approach of 96.40% and 96.59% respectively. Bandwidth approach obtained 5.49% higher classification rate than 3-formant features using MLP. 2009-08-13T03:00:40Z 2009-08-13T03:00:40Z 2009-07 Article p.75-79 978-1-4244-1723-0 http://ieeexplore.ieee.org/xpls/abs_all.jsp?=&arnumber=4590133 http://hdl.handle.net/123456789/6869 en Proceedings of the International Conference on Audio, Language and Image Processing (ICALIP 08) Institute of Eelectrical and Electronics Engineering (IEEE)
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 Feature extraction
Speech recognition
Natural language processing
Multilayer perceptrons
Digital electronics
Signal theory (Telecommunication)
Automatic speech recognition
Speech perception
spellingShingle Feature extraction
Speech recognition
Natural language processing
Multilayer perceptrons
Digital electronics
Signal theory (Telecommunication)
Automatic speech recognition
Speech perception
Paulraj, M. P.
Sazali, Yaacob
Mohd Yusof, S. A.
Vowel recognition based on frequency ranges determined by bandwidth approach
description Link to publisher's homepage at http://ieeexplore.ieee.org
format Article
author Paulraj, M. P.
Sazali, Yaacob
Mohd Yusof, S. A.
author_facet Paulraj, M. P.
Sazali, Yaacob
Mohd Yusof, S. A.
author_sort Paulraj, M. P.
title Vowel recognition based on frequency ranges determined by bandwidth approach
title_short Vowel recognition based on frequency ranges determined by bandwidth approach
title_full Vowel recognition based on frequency ranges determined by bandwidth approach
title_fullStr Vowel recognition based on frequency ranges determined by bandwidth approach
title_full_unstemmed Vowel recognition based on frequency ranges determined by bandwidth approach
title_sort vowel recognition based on frequency ranges determined by bandwidth approach
publisher Institute of Eelectrical and Electronics Engineering (IEEE)
publishDate 2009
url http://dspace.unimap.edu.my/xmlui/handle/123456789/6869
_version_ 1643788614305316864
score 13.222552

Similar Items