Dynamic sound rendering for distributed virtual environments

Abstract

Dynamic sound rendering is attracting a lot of attention in recent years due to its applications in computer games and architecture simulation. Although physical based methods can produce realistic outputs, generic physical based methods typically involve recursive ray-tracing, which may not be applicable in interactive dynamic environments. In this thesis, we propose a distributed sound rendering architecture for real-time sound rendering in an interactive environment with moving observers and sound sources. The new architecture is based on low cost PCs and is scalable. For the distributed architecture, we propose two methods to accelerate the sound rendering process: the distributed prioritized sound rendering method, and the ray caching method. The distributed prioritized sound rendering method applies prioritization technique on the distributed environment to improve the sound rendering performance while preserving the perceptual quality. The ray caching method exploits ray coherency to accelerate the ray-tracing process. It is tailored for interactive sound rendering based on two approximation techniques: spatial approximation and angular approximation. The ray cache supports intra-frame, inter-frame and inter-observer sharing of projected rays. We have implemented the proposed methods in a prototype sound rendering system. We show the performance of the new methods through a number of experiments and discuss the performance tuning issues.