한국 가수들의 창법 및 음색의 디지털적 특징 분석

Abstract

This study pursues the possibility of sound visualization tool, called spectrogram which best enables quantitative as well as objective analysis of sound quality. In order to effectively implement the application of spectrogram four representative Korean singers are selected: two trot singers Lee Me-ja & Joo, Hyun-me; one operatic singer, Shin, Young-ock and one new folk style singer, Yang, Hee-eun. The selection is based on the differences in their singing techniques, vibration types, scope and density of their voice types all of which determine the sound quality, timbre of the singers. This kind of quantitative analysis of sound makes it easy to devise some sound modulation tool to transform amateur's singing into specific professional's singing quality and thus eventually enables inventing a new imitation Noraebang machine or microphone with sound modulator usable in any fun stage. Or it could analyze one's singing habit and correct some bad singing habit into more desired one, serving as a vocal training tool to visually enhance one's singing quality. And with accumulated database of singing techniques and timbre characteristics of various singers, sound synthesis through signal processing techniques can freely create hitherto-unknown voice qualities.