Issues and Challenges in Sindhi Speech Recognition Syste

T. QURESHI, D.N. HAKRO, F. H. KHOSO, K.U.R. KHOUMBATI, R. TAKHARWAL, M. HAMEED, G. N. RAJPER

Abstract


Speech Recognition System is the conversion of audio signals of human to computer understandable signals. These understandable signals are then converted into editable writing of text on the screen or these signals can also be used to control a computer device. The recognition mechanism is also possible on mobile devices where a user speaks to a smartphone device and smartphone understands the speech signals. Much of the research work is available on various aspects of Sindhi language and Sindhi computing as Sindhi Language has a rich history of more than 5000 years. Speech recognition research studies on many of the languages are found easily but no any work has been found on Sindhi language. These languages possessing SR systems are mostly of non-cursive nature containing a smaller number of peculiarities. Establishing a speech recognition system of Sindhi language can be a cumbersome and challenging task. This study presents a detail of issues and challenges in development of Sindhi speech recognition system. The main challenges in development of Sindhi speech recognition have been discussed are various types of consonants, phonetic properties, consonants phonemes, sound variation according to the dots, discretization, noise, environment, speaker, gender speech and others. Recognition of native speakers and non-native speakers is a challenging task due to the many accents available in Sindhi


Full Text:

PDF

Refbacks

  • There are currently no refbacks.


Copyright (c) 2018 Sindh University Research Journal - SURJ (Science Series)

 Copyright © University of Sindh, Jamshoro. 2017 All Rights Reserved.
Printing and Publication by: Sindh University Press.