The Recognition of Holy Qur’an Reciters Using the MFCCs’ Technique and Deep Learning

Abstract

The Holy Qur’an has recently gained recognition in the field of speech-processing research. It is the central book of Islam, from which Muslims derive their religious teachings. The Qur’an is the primary source and highest authority for all Islamic beliefs and legislation. It is also one of the most widely memorized and recited texts around the world. Listening to and reciting the Qur’an is one of the most important daily practices for Muslims. In this study, we propose a deep learning model using convolutional neural networks (CNNs) and a dataset consisting of seven well-known reciters. We utilize mel frequency cepstral coefficients (MFCCs) to extract and evaluate information from audio sources. We compare our proposed model to different deep learning and machine learning methodologies. Our proposed model outperformed the competing models with an accuracy of 99.66%, compared to the support vector machine’s accuracy of 99%.