The Study of Mapping Strategies Between the Excitators of the Single-Reed Woodwind and the Bowed String

摘要

Mapping is one of the most important components in digital musical instruments. There have been many works on "longitudinal" mapping strategies from the input device to the sound synthesizer. Such mapping can be considered longitudinal because the mapping direction is in line with the information transfer direction. However, less research focuses on "transversal" mappings among input devices or sound synthesizers. In this paper, a transversal mapping strategy is explored between the excitators of bowed strings and single-reed woodwinds which aims to allow more natural use of a given controller to play the sound of another family of instruments. A three-layer mapping structure, namely the playing layer, the mathematical layer, and the physical layer, is built. The mappings in different layers are generated based on the analogy of the mathematical models of two excitators in the mathematical layer. As a result, in the playing layer, the bowing force and the bowing speed of the string instrument are mapped to the lip force and the mouth pressure of a single-reed woodwind, respectively. In the physical layer, the string velocity and the friction force at the bowed point are mapped to the acoustic pressure and the volume velocity in the mouthpiece, respectively. Finally, a Yamaha WX5 wind controller is used to drive the digital waveguide string model. Two different mapping strategies between the lip force and the bowing force are tested and the results are discussed.