Spatially correlated audio in multipoint videoconferencing

摘要

The disclosed method provides audio location perception to an endpoint in a multipoint videoconference by providing a plurality of audio streams to the endpoint, wherein each of the audio streams corresponds to one of the loudspeakers (211 a-n) at the endpoint (210a-n) . The audio streams are differentiated so as to emphasize broadcasting of the audio streams through one or more loudspeakers closest to a position of a speaking endpoint in a videoconference layout that is displayed at the endpoint. For example, the audio broadcast at a loudspeaker that is at a far-side of the screen might be attenuated or time delayed compared to audio broadcast at a loudspeaker that is located at a near-side of the display. The disclosure also provides a multipoint control unit (MCU) that processes audio signals from two or more endpoints according to the positions in a layout of the endpoints and then transmits processed audio streams to the endpoints in a way that allows endpoints to broadcast spatially correlated audio.