Physically based models for liquid sounds

A physically based liquid sound synthesis methodology is developed. The fundamental mechanism for the production of liquid sounds is identified as the acoustic emission of bubbles. After reviewing the physics of vibrating bubbles as it is relevant to audio synthesis, a sound model for isolated single bubbles is developed and validated with a small user study. A stochastic model for the real-time interactive synthesis of complex liquid sounds such as produced by streams, pouring water, rivers, rain, and breaking waves is based on the synthesis of single bubble sounds. It is shown how realistic complex high dimensional sound spaces can be synthesized in this manner.