Cover Image

Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?

This paper explains works in speech recognition using neural network. The main objective of the experiment is to choose suitable number of nodes in hidden layer and learning parameters for malay iIsolated digit speech problem through trial and error method. The network used in the experiment is feed...

Full description

Saved in:
Bibliographic Details
Main Authors: Salam, Md. Sah, Mohamad, Dzulkifli, Salleh, Sheikh
Format: Article
Published: Zarka Private University, Colleges of Computing and Information Technology 2011
Subjects:
BF Psychology
Online Access:http://eprints.utm.my/id/eprint/45025/
http://ccis2k.org/iajit/PDF/vol.8,no.4/4-969.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.utm.45025
record_format eprints
spelling my.utm.450252019-03-07T00:59:35Z http://eprints.utm.my/id/eprint/45025/ Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters? Salam, Md. Sah Mohamad, Dzulkifli Salleh, Sheikh BF Psychology This paper explains works in speech recognition using neural network. The main objective of the experiment is to choose suitable number of nodes in hidden layer and learning parameters for malay iIsolated digit speech problem through trial and error method. The network used in the experiment is feed forward multilayer perceptron trained with back propagation scheme. Speech data for the study are analyzed using linear predictive coding and log area ratio to represent speech signal for every 20ms through a fixed overlapped windows. The neural network learning operation is greatly influenced by the parameters i.e., momentum, learning rate and number of hidden nodes chosen. The result shows that choosing unsuitable parameters lead to unlearned network while some good parameters set from previous work perform badly in this application. Best recognition rate achieved was 95% using network topology of input nodes, hidden nodes and output nodes of size 320:45:4 respectively while the best momentum rate and learning rate in the experiment were 0.5 and 0.75. Zarka Private University, Colleges of Computing and Information Technology 2011 Article PeerReviewed Salam, Md. Sah and Mohamad, Dzulkifli and Salleh, Sheikh (2011) Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters? International Arab Journal of Information Technology, 8 (4). pp. 364-371. ISSN 1683-3198 http://ccis2k.org/iajit/PDF/vol.8,no.4/4-969.pdf
institution Universiti Teknologi Malaysia
building UTM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
url_provider http://eprints.utm.my/
topic BF Psychology
spellingShingle BF Psychology
Salam, Md. Sah
Mohamad, Dzulkifli
Salleh, Sheikh
Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
description This paper explains works in speech recognition using neural network. The main objective of the experiment is to choose suitable number of nodes in hidden layer and learning parameters for malay iIsolated digit speech problem through trial and error method. The network used in the experiment is feed forward multilayer perceptron trained with back propagation scheme. Speech data for the study are analyzed using linear predictive coding and log area ratio to represent speech signal for every 20ms through a fixed overlapped windows. The neural network learning operation is greatly influenced by the parameters i.e., momentum, learning rate and number of hidden nodes chosen. The result shows that choosing unsuitable parameters lead to unlearned network while some good parameters set from previous work perform badly in this application. Best recognition rate achieved was 95% using network topology of input nodes, hidden nodes and output nodes of size 320:45:4 respectively while the best momentum rate and learning rate in the experiment were 0.5 and 0.75.
format Article
author Salam, Md. Sah
Mohamad, Dzulkifli
Salleh, Sheikh
author_facet Salam, Md. Sah
Mohamad, Dzulkifli
Salleh, Sheikh
author_sort Salam, Md. Sah
title Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
title_short Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
title_full Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
title_fullStr Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
title_full_unstemmed Malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
title_sort malay isolated speech recognition using neural network: a work in finding number of hidden nodes and learning parameters?
publisher Zarka Private University, Colleges of Computing and Information Technology
publishDate 2011
url http://eprints.utm.my/id/eprint/45025/
http://ccis2k.org/iajit/PDF/vol.8,no.4/4-969.pdf
_version_ 1643651619627204608
score 13.211869

Similar Items