Universiti Teknologi Malaysia Institutional Repository

Speaker-independent Malay vowel recognition of children using multi-layer perceptron

Hua, Nong Ting and Yunus, Jasmy (2004) Speaker-independent Malay vowel recognition of children using multi-layer perceptron. In: IEEE Region 10 Annual International Conference, Proceedings/TENCON. TENCON-IEEE Region 10 Conference Proceedings, A . IEEE, USA, pp. 68-71. ISBN 0-7803-8560-8

[img] PDF
Restricted to Repository staff only

1MB

Official URL: http://dx.doi.org/10.1109/TENCON.2004.1414358

Abstract

Most of the speech recognitions are based on adult speech sounds. Less research is done in the recognition of children speech sounds. The speech of children is more dynamic and inconsistent if compared to adults speech. This paper investigates the use of neural networks in recognizing 6 Malay vowels of Malay children in a speaker-independent manner. Multi-layer Perceptron with one hidden layer was used to recognize these vowels. The Multi-layer Perceptron was trained and tested with speech samples of Malay children with their ages between seven and ten years old. A single frame of cepstral coefficients were extracted around the vowel onset point using Linear Predictive Coding. The vowel length was examined from 5 ms to 70 ms. Experiments were conducted to determine the optimal vowel length as well as the number of cepstral coefficients.

Item Type:Book Section
Additional Information:ISBN: 0-7803-8560-8 ; IEEE Region 10 Conference on Analog and Digital Techniques in Electrical Engineering; 21-24 Nov. 2004, Chiang Mai, Thailand.
Uncontrolled Keywords:multilayers, neural networks, speech coding, linear predictive coding, multi-layer perceptrons, vowel length, speech recognition
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:Electrical Engineering
ID Code:9823
Deposited By: Zalinda Shuratman
Deposited On:29 Mar 2010 02:28
Last Modified:15 Aug 2017 04:18

Repository Staff Only: item control page