Quranic verses verification using speech recognition techniques

Abstract

Al-Quran is the holy book of Muslims which is written and recited in Arabic language, the language in which it was revealed. Muslims believe that the Quran is neither corrupted nor altered this is mainly due to maintaining its original text. It is forbidden to recite the Quran in any other language apart from Arabic with neither additions nor subtractions. However with the proliferation of technology especially the Internet and social media sites such as Facebook and Twitter, the spread of mistakenly or deliberately distorted audio clips are witnessed regularly. In this regard, it is necessary to preserve the authenticity and integrity of the Quran from all sorts of corruption. This paper describes challenges and solutions for building a successful verification system of the Quran verses online. The paper describes the techniques used to deal with a finite vocabulary how modelling completely in the voice domain for language model and dictionary can avoid some system complexity, and how we built dictionaries, language and acoustic models in the framework.