PENGGOLONGAN SUARA BERDASARKAN USIA DENGAN MENGGUNAKAN METODE K - MEANS

Speech technology is one of the technological applications that have been found several years ago. One is the speaker recognition a process which is often referred to as verification speakers. Which means recognizing a voice by comparing with the standard. voice feature extraction can be performed by the sampling process, front-end, preemphasis, frame blocking, windowing, and DFT (Discrete Fourier Transform) from DFT to obtain the first formant and formant 2 and then processed with K - Means and results of K - Means stored in the database. if incoming speech signal, the system will perform speech processing. Then a new signal extraction results will be compared to a standard signal extraction results that is in the database so that will be compared with results clustering, whether the vote was within the range centroid 1 (adult) or centroid 2 (children). Results of this software is a sound clustering adult and children, which will differentiate the system will sound an adult and children by looking at his second formant values. Where to cluster red circle its adult and blue circles its children. Keywords: speaker recognition verification of speakers, speech recognition, grouping types of age, Sampling, Front End Detection, Blocking Frame, Pre - emphasis, Windowing, DFT, K - Means