Analysis of a distributed processing model for spatialized audio conferences

New technologies such as the IP multimedia subsystem (IMS) provide components to support enhanced audio conferencing services, requiring media distribution models designed for traditional monophonic VoIP audio conferences to be revised and extended. We propose a new media distribution model, for headphone based spatialized audio conferences where spatial cues are provided by head related transfer function (HRTF) filtering. Under our proposal, the HRTF filtering and mixing processes may be partially distributed across conferees terminals and a dedicated audio processing server. This results in a high level of spatial manipulation per terminal, whilst server processing load is minimized. We also provide analysis of two network load reduction techniques for our model; application layer routing (ALR) and optimized mix selection (OMS), demonstrating significant load reduction when using ALR for conferences ranging in both size, and terminal processing capabilities.