Biometric Information Based On Distribution of Arabic Letters According To Their Outlet

Abstract

In this research, the correlation and homogeneity properties of the presence matrix of the speech signal for the Arabic letters were tested and evaluate the possibility of distinguishing between them was achieved by extracting characteristic properties. The speech signal of the letters (acquired through the Recorder) with a binary matrix for its configuration and calculations of two properties of presence matrix, correlation, homogeneity studied. The values of these properties and the extent of their variation from one person to another and how close they are within the same group represent the exits which they belong. The results in this paper illustrate the correlation and homogeneity properties of the Arabic letters to persons in alphabetical order provide a distinctive description for the person. The important of Arabic language or any other language how they affected by an significant factors as the economic situation of their users and civilization, as well as their scientific future. This study is a good and authentic attempt in terms of using two types of relationships (correlation and homogeneity properties) , whose results will lead to important conclusions in the field of extracting the properties of sound and its difference from one person to another, depending on the outlets of speech. The results of the practical application for the proposed algorithm show that correlation gave distinct results for the adoption of characters as a characteristic of the voice of the speaker, whereas homogeneity was a weak indicator that varied greatly for the same character with the same person.