摘要:
Teleconferencing and, in particular, distributed teleconferencing may use methods and systems for location grouping to reduce feedback and other audio anomalies. Terminals and users connected to the same teleconference and in the same location might not need to receive audio signals from the other terminals and users in the same location. As such, by detecting and analyzing the location of each participating terminal, the terminals (and thus, the users thereof) may be organized into location groups to provide proper audio mixing. In one example, first and second terminals in the same location might not receive each other's audio in a downstream teleconference signal. The location and grouping of terminals may be processed using context fingerprint information derived from sensor readings of each terminal. Sensors may include GPS sensors, cameras, BLUETOOTH sensors and the like. Context fingerprint information may further be synchronized to enhance location determination and grouping.
摘要:
Provided are multichannel architectures, systems, methods, and computer program products for distributed teleconferencing using one or more master devices and/or a centralized conferencing switch. Multichannels enhance functionality of a master device in distributed teleconferencing and allow for compatibility with 3D capable teleconferencing. Multichannel distributed teleconferencing involves multichannel, monophonic, and/or a fixed number of uplink and downlink channels. A multichannel distributed teleconferencing system may perform active talker detection of near-end participants and communicate an ID signal on an uplink channel identifying the active near-end participants. A multichannel distributed teleconferencing system may also receive an ID signal on a downlink channel identifying the active far-end participants. A multichannel distributed teleconferencing system may perform various uplink and downlink processing. Uplink processing may involve multimixing and spatialization. Multimixing may be used to separate speech signals of near-end participants. Spatialization, also used in downlink processing, introduces spatial separation of active participants.