International Journal on Islamic Applications in Computer Science And Technology, Vol. 4, Issue 2, June 2016, xx-xx Towards Using CMU Sphinx Tools for the Holy Quran Recitation Verification Mohamed Yassine El Amrani 1,2,a , M. M. Hafizur Rahman 2,b , Mohamed Ridza Wahiddin 2,c , and Asadullah Shah 2,d 1 Jubail University College, Saudi Arabia 2 International Islamic University Malaysia, Malaysia a amranim@ucj.edu.sa, b hafizur@iium.edu.my, c mridza@iium.edu.my, d asadullah@iium.edu.my ABSTRACT The use of the Automatic Speech Recognition (ASR) technology is being used in many different applications that help simplify the interaction with a wider range of devices. This paper investigates the use of a simplified set of phonemes in an ASR system applied to Holy Quran. The Carnegie Mellon University Sphinx 4 tools were used to train and evaluate a language model on Holy Quran recitations that are widely available online. The building of the language model was done using a simplified list of phonemes instead of the mainly used Romanized in order to simplify the process of training the acoustic model. In this paper, the experiments resulted in Word Error Rates (WER) as low as 1.5% even with a very small set of audio files used during the training phase. Index Terms—Automatic speech recognition, Holy Quran recognition, Human voice. 1 INTRODUCTION The Holy Quran is at the center of every Muslim in the world. Practicing 5 daily prayers is one of the five pillars of Islam and every prayer involves some Holy Quran recitation. Hence, every Muslim is involved in some Holy Quran memorization in order to recite some Ayat (“sentences” in the Holy Quran) during his/her daily prayers. Many Ayat encourage memorizing the Holy Quran and the mention of its easiness of memorization is repeated several times as in the Ayah: { ر ك͉Ϊ ͊ ن م مْ لϬ ف رْ ك͋ άϠ لϥ آْ ر Ϙ ْ الΎ نْ ر͉δ يْΪϘ ل ϭ} “And We have certainly made the Quran easy for remembrance, so is there any who will remember?” (50:17) As of any memorization process, a continuous review is required to remember as what was previously memorized. The best way to have one’s recitation reviewed and verified is to involve another person who follows one’s recitation and corrects his mistakes. When no one is available to check someone’s recitation, the verification cannot be done easily since the reciter has to continuously check after each group of Ayat if his recitation was correct. In this case, the reciter