Arabic phonemes recognition system based on malay speakers using neural network

Author

Abd Almisreb, Ali ; Abidin, Ahmad Farid ; Md Tahir, Nooritawati

Author_Institution

Fac. of Electr. Eng., Univ. Teknol. MARA, Shah Alam, Malaysia

fYear

2014

fDate

Sept. 28 2014-Oct. 1 2014

Firstpage

188

Lastpage

192

Abstract

Arabic language can be used by native and nonnative speakers; due to Arabic is the language of the holy book of Muslims. In this paper, Arabic phoneme recognition system is proposed based on Malay speakers. This system consists of three main stages. The first stage is noise reduction and it aims to enhance the phoneme signals by excluding the unvoiced signals and keep only the voiced signal. Wiener filter is adapted to accomplish this task. The second stage is based on Mel-Frequency Cepstral Coefficients method to extract a vector of features to represent each phoneme signal. Eventually, pattern recognition neural network is designed as recognizer. The proposed system produces sufficient outcomes with 20 hidden neurons.

Keywords

Wiener filters; filtering theory; natural language processing; neural nets; signal representation; speaker recognition; Arabic language; Arabic phonemes recognition system; Malay speakers; Muslims; Wiener filter; feature vector extraction; mel-frequency cepstral coefficients method; native speakers; noise reduction; nonnative speakers; pattern recognition neural network; phoneme signal enhancement; phoneme signal representation; Artificial neural networks; Feature extraction; Speech; Speech recognition; Training; Wiener filters; Arabic; Malay; Mel-Frequency Cepstral Coefficients; Wiener; pattern recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Wireless Technology and Applications (ISWTA), 2014 IEEE Symposium on

Conference_Location

Kota Kinabalu

Print_ISBN

978-1-4799-5435-3

Type

conf

DOI

10.1109/ISWTA.2014.6981184

Filename

6981184