Analysis of Formant Frequencies of the Correct Pronunciation of Quranic Alphabets Between Kids and Adults

Abstract

It is an obligation for a Muslim to become skilled and proficient in reciting Al-Quran considering that Al-Quran is the fundamental source of revelation from Allah SWT. In Al-Quran, there are 28 alphabets where each of them has their own unique sound. The Quranic alphabets produce sound that are characterized from their point of articulation (Makhraj) and their characteristics (Sifaat). Knowing the correct way of pronunciation through engineering perspective may help Muslim in learning Al-Quran, in the sense that the signal of the experts can be used in Quranic teaching and learning as a reference model. Since both adults and children possess different vocal tract, therefore there will be different outcomes of the pronunciation between both experts. The features identification of the pronunciation of both experts is needed to represent the actual and correct pronunciation that will be used as a reference for Quranic teaching and learning at later. In this paper, the focus was on the identification and analysis of the correct pronunciation of the Quranic alphabets on the data obtained from adults and children experts. The first and second formant frequencies (F1 and F2) were used as the features where they were used to represent the pronunciation of each alphabet for both adults and children category. The speech analysis software PRAAT was used to accomplish the pre-processing of the data using Spectral Subtraction technique and also used to measure the F1 and F2 values. Linear Discriminant Analysis (LDA) was used for classification of the signals and results shows that some of the alphabets can be identified uniquely using F1 and F2 features of the two categories.