Features Identification and Classification of Alphabet ? (ro) in Leaning (Al-Inhiraf) and Repetition (Al-Takrir) Characteristics

摘要

It is important for Muslim to recite the Quran properly with the correct Tajweed. which includes the use of correct characteristics (sifaat) and point of articulations (makhraj). To this date, there are limited researches done focusing on classifying the Quranic letters according to the characteristics. In this study, the focus is given to the classification of the characteristics of the Quranic letters for the purpose of developing an automated self-learning system for supporting the conventional method of Quranic teaching and learning. The characteristics of Quranic letters, which are the focus in this paper are Leaning and Repeating, where both consists of ? (ro) alphabet. Several methods of feature extractions and analysis were implemented such as Formant Analysis, Power Spectral Density (PSD), and Mel Frequency Cepstral Coefficient (MFCC) to come out with the suitable features that best represent the correct characteristics of the alphabet. Once the features had been identified, Linear Discriminant Analysis (LDA) and Quadratic Discriminant Analysis (QDA) were used as the classifier. The results show that QDA with all 19 features trained achieved the highest percentage accuracy for both Leaning ( Al-Inhiraf) and Repetition (Al-Takrir) characteristics with of 82.1% and 95.8% of accuracy respectively.