Development of synthesis model for fine tuning and playing style

Abstract

This paper describes a synthesis model that is able to represent playing style and fine tuning for the plucked string instrument. The proposed model is based on digital waveguide theory and the target instrument is the Gayageum, traditional Korean plucked string instrument. It has 12 twisted silk strings, each supported by a movable bridge called Anjok. Each of them is placed on the body; it tunes each string by being moved sideways along a string. Therefore, the movable bridge model is proposed to determine pitch of synthetic sound; this is equivalent to determine length of a delay line in string model of digital waveguide model. The fundamental frequency of each string corresponding to the location of the Anjok was measured and parameterized. This parameter is also used in the string model to compute the loop filter coefficient. Finally, the unique right-hand playing style, flicking with the middle and index fingers, is represented by a FIR filter. Although there were differences between the original and synthesized sound by proposed synthesis model, the entire harmonic magnitude envelope of the synthesized sound followed those of the original sound and it was aurally hard to distinguish them.