A new system for Arabic recitation using speech recognition and Jaro Winkler algorithm

Abstract

Automated recitation plays an important role in improving self-learning. There is a limited number of applications for Arabic automated recitation focusing only on Holy Qur’an. This article proposed a new system (Samee’a – سميع ) to facilitate memorizing any kind of text such that poem, speeches and many more. Samee’a system is based on Google Cloud Speech Recognition API and Jaro Winkler Distance algorithm that determines the similarity between the original and recited text. The system has been tested on over 70 files ranging between 12 to 400 words. The average similarity achieved 85.6%. To validate the obtained results, a comparison study was successfully performed on some chapters of the Holy Qur’an. Also, the user experience testing was carried out by 10 users of different ages (between 5 and 50-year-old) using small texts and some chapters of Holy Qur’an. The proposed system proved its efficiency.