Research On Audio Of Virtual Space Teleconferencing System

Virtual Space Teleconferencing (VST) system is developed on the basis of the multimedia conferencing system. All participants are substituted with virtual humen in the VST system generated with computer. Objects in the VST system can be positioned, observed and manipulated by the participants with their avatars. Besides that, avatars are also used to achieve "face to face" communication among participants who take part in the VST. Audio is both the main carrier and the way of communication in VST system. Therefore, the audio technology plays a vital role on the development of VST system.The VST system is developed with virtual simulation software Vega and DI-Guy. In this paper, the audio technology of VST is studied intensively. Some solutions of the audio technology are presented, such as the real-time acquisition for audio data, the encoding and compressing method, the transmission for audio data stream in network, the decoding and playing method in the virtual space, and so on. The solutions mentioned before are developed and implemented with the VC++ and Multimedia Audio SDK. And the VST system is debugged and run successfully in Windows. In addition, in order to enhance the immersion of the VST system, the subject of Lip synchronization process for virtual avatars is studied further, and the specific solutions are presented.