Efficient Description and Rendering of Complex Interactive Acoustic Scenes

Interactive environmental audio spatialization technology has become commonplace in personal computers and is migrating into portable entertainment platforms (including cell phones) and multiplayer game servers (virtual online worlds). While the primary current application of this technology is 3D game sound track rendering, it is ultimately necessary in the implementation of any personal or shared immersive virtual world (“virtual reality”). The successful development and deployment of such applications in new mobile or online platforms involves maximizing the plausibility of the synthetic 3D audio scene while minimizing the computational and memory footprint of the audio rendering engine. It also requires a flexible, standardized scene description model to facilate the development of applications targeting multiple platforms. This paper reviews a computationally efficient 3-D positional audio and spatial reverberation processing architecture for real-time virtual acoustics over headphones or loudspeakers, compatible with current interactive audio standards (including MPEG-4, OpenAL, JSR 234 and OpenSL ES).