Mapping and Clustering of the Acoustic Signal of Malay Traditional Musical Instruments using Statistical Analysis Method

This paper discusses the development of a method for mapping and clustering acoustic signal of traditional malay musical instruments. the acoustic signal for knot b3 has been recorded using a sound recording system which has been designed based on the combination of a microphone, a power module and a data acquisition system. the malay traditional musical instruments involved are cak lempong, flute and gambus. the recording in the time domain shows a variety of waveforms according to the instrument’s method of play. the level of similarity between two different acoustic signals of same traditional malay musical instrument has been determined by anova analysis. for the mapping and clustering process, alternative methods of statistical signal analysis known as integrated kurtosis-based algorithm for z-notch filter (i-kaz™) and mesokurtosis zonal nonparametric (m-z-n) has been applied to extract the signal’s features. identification is done using a combination of i-kaz and m-z-n coefficients. the result shows that the characterization of acoustic signals of traditional malay musical instruments through mapping and clustering has been successfully implemented by this developed method.