Analisis Discrete Sine Transform (DST) dan Discrete Cosine Transform (DCT) pada Suara Sandi Morse

Abstract

Voice recognition is a process to recognize the characteristics of a person's voice to recognize the speech of the person. The beginning stage a make a speech recognation system required sampling process. Interest voice sampling is 10 obtain the characteristics of the voice to be used as a reference pattern. Sampling used in this study is the voice sampling Morse code that produced through the aid of a whistle. Morse code the speech recognition system is built using Delphi XE by using discrete sine transform (DST) and discrete cosine transform (DCT). In this study, the authors wanted to see the degree to akurasian of methods of discrete sine transform (DST) and discrete cosine transform (DCT) in the process of voice recognition to translate Morse code in the form of letters of the alphabet pattern. The reason for choosing the method of discrete sine transform (DSI) and discrete cosine transform (DCT) in the voice recognition Morse code is due to the discrete sine transform (DST) and discrete cosine transform (DCT) has the same advantages in the calculation process which is able to simplify the process and reduce complexity $0 itis suitable for signal processing and sound. 1" just for discrete sine transform (DST) in the process of calculating the sampling sounds using sin function while the discrete cosine transforms (DCT) receipts cos function. The performance of discrete sines transform (DST) and discrete cosine transform (DCT) 1s evaluated based on two parameters: rue detection and false detection. The results showed that the average successful recognizable voice samples of discrete sine transform (DST) and discrete cosine transform (DCT) Whereas 86.5% of the sample were not successfully identified only 13.5%,