文章基本信息

标题：Sindhi Speech Recognition System
本地全文：下载
作者：Fida Hussain Khoso ; Dil Nawaz Hakro ; Syed Zafar Nasir 等
期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2019
卷号：19
期号：11
页码：21-28
出版社：International Journal of Computer Science and Network Security
摘要：For languages around the world, one of the most revolutionary and significant technology is the Speech Recognition, which helps in advancement of languages computing. Speech recognition systems for most of the languages around the world have either been already developed or under the process of development. Unlike other languages, the number of characters and the sounds in Sindhi language is more than the other languages, which makes it unique so most of the available approaches cannot be used for Sindhi language. Therefore, a totally different approach is required in correspondence of Sindhi sounds and characters for designing and development of the tools for Sindhi Speech Recognition. The literature review of Sindhi speech recognition reveals that there is a gap in the quality work and no SR system is available (to the best of our knowledge) with respect to software development for recognition of spoken Sindhi characters and sounds, which may be converted in Sindhi writing by the software. If the input is made through Sindhi speaking, it will not only help the people in reducing effort of typing of very difficult character sounds and characters of Sindhi but also be beneficial for developers who need to map the 52 characters of Sindhi Alphabets against 26 alphabets of English keyboard.
关键词：Speech Recognition; Acoustic Model; Artificial Intelligence; Natural Language Processing.