Development and Research of a System for Automatic Recognition of the Digits Yemeni Dialect of Arabic Speech Using Neural Networks

Author:

Radan N.H.1ORCID,Sidorov K.1ORCID

Affiliation:

1. Tver State Technical University

Abstract

The article describes the results of research on the development and testing of an automatic speech recognition system (SAR) in Arabic numerals using artificial neural networks. Sound recordings (speech signals) of the Arabic Yemeni dialect recorded in the Republic of Yemen were used for the research. SAR is an isolated system of recognition of whole words, it is implemented in two modes: "speaker-dependent system" (the same speakers are used for training and testing the system) and "speaker-independent system" (the speakers used for training the system differ from those used for testing it). In the process of speech recognition, the speech signal is cleared of noise using filters, then the signal is pre-localized, processed and analyzed by the Hamming window (a time alignment algorithm is used to compensate for differences in pronunciation). Informative features are extracted from the speech signal using mel-frequency cepstral coefficients. The developed SAR provides high accuracy of the recognition of Arabic numerals of the Yemeni dialect – 96.2 % (for a speaker-dependent system) and 98.8 % (for a speaker-independent system).

Publisher

Bonch-Bruevich State University of Telecommunications

Reference12 articles.

1. Al-Zabibi M. An acoustic-phonetic approach in automatic Arabic speech recognition. Loughborough University. Doctoral Thesis. 1990. URL: https://hdl.handle.net/2134/6949 [Accessed 02.10.2023]

2. Alkhouli M. Alaswaat Alaghawaiyah. Daar Alfalah, Jordan. 1990 (in Arabic)

3. Deller J., Hansen J., Proakis J. Discrete-Time Processing of Speech Signal. 1993. DOI:10.1109/9780470544402

4. Elshafei M. Toward an Arabic Text-to-Speech System. The Arabian Journal for Science and Engineering. 1991;16(4B):565‒583.

5. Hagos E. Implementation of an Isolated Word Recognition System. M.Sc. Thesis. King Fahd University of Petroleum & Minerals Dhahran, Saudi Arabia. 1985.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3