Developing an application that detects vocal errors in the recitation of the holy Quran

Abstract

The goal of this thesis is to facilitate the process of receiving the Noble Qur’an by creating a program that detects voice errors for the Qur’an reciter, this was done by means of a “web speech recognizer” which recognizes the reciter's reading as the first step and then converts it into a visible text for the user via the Speech-To-Text process. This text is compared with the original text of the Noble Qur’an which has been downloaded in the form of a database, after full compatibility between the two texts, the user is notified of his correct reading. In the event of a mistake in user recitation, the user's text differs from the original text, then he is notified of his incorrect reading. After creating a web-based application, we searched for similar works. We found three works: “Tamkeen”, “Read Along” Mobile Applications and “Web captioner”. Each of them has its own goal, but they meet our application in at least one of its goals. Finally, in this thesis, we experimented the effectiveness of our application through its 20 experiments, saving the results of these experiments, discussing them, and giving possible reasons for these results with proposing future solutions.

Key words: The Holy Quran, Speech-To-Text process, Arabic language, web-based application.