Implementation of Dynamic Time Warping algorithm on an Android based application to write and pronounce Hijaiyah letters

Abstract

Understanding Hijaiyah letters is one of the important things for Muslims to be able to read the Qur'an. Dynamic Time Warping algorithm can be used to study similarity on writing and pronunciation for Android users. Dynamic Time Warping algorithm used to calculate the ratio of existing data and data input from the user to see the resemblance. In the calculation, Dynamic Time Warping Algorithm obtains image data from the extracted images and sound using two methods. To process the image data Principle Component Analysis is used and for the voice data processing Mel Frequency Cepstrum Coefficients method is used. From the calculation of the data, it can be seen the similarities on writing and pronunciation of users who will be raised in the form of value. For comparison accuracy Euclidean Distance method is used. From this research can be found Dynamic Time Warping algorithm can be implemented to determine the similarity of the writing and pronunciation of letters hijaiyah.