تفاصيل الرسالة العلمية

Although the Quranic passages were taken especially from the same Quranic verse, the method of reading the sentence in the Quran in terms of recitation may be different due to the application of the provisions of Tajweed. It may result in different voices for different readers. In this research, we try to overcome the above challenges by developing a complete voice recognition system that extracts the holly Quran Tajweed rules from a given Quran Suras. The proposed system consists of three stages: the preprocessing stage, the feature extraction stage and feature classification and pattern recognition stage. In the preprocessing stage, we implement the endpoint detection, pre-emphasis filtering, noise filtering, and channel normalization. In the second stage, features are extracted using the Mel-Frequency Cepstral Coefficients (MFCC). Finally, the classification stage is carried out using three classifiers, Hidden Markov Model (HMM), Artificial Neural Network (ANN) and Support Vector Machine (SVM). The data set consists of the voice of three Quran Suras (Elnas, Fatihah, Ikhlas). These suras are inserted using the voice of different readers. The proposed system was tested using 20, 60 and 90 audio files. The best results are achieved when we used the ANN as a classifier. We have reached 98.9 % Recognition Rate