Selective audio combination for a conference

A method for performing a conference, comprising:

• a multipoint control unit (MCU) hosting a conference between a plurality of participants at respective participant locations, wherein the conference is performed using a plurality of endpoints at the respective participant locations;
• the MCU receiving audio from each of the plurality of videoconferencing endpoints;
• the MCU receiving audio level information from each of the plurality of endpoints;
• the MCU combining the audio from a plural subset of the plurality of endpoints to produce conference audio, wherein the audio from the plural subset is combined based on the audio level information, wherein the plural subset is less than all of the plurality of conferencing endpoints, wherein said combining comprises analyzing the audio level information from each of the plurality of endpoints to determine audio from a designated number (n) of videoconferencing endpoints which have highest audio levels, wherein the value of n is determined dynamically based on said analyzing, and wherein said combining the audio from the plural subset of the plurality of endpoints comprises combining only the audio from the n endpoints;
• and
• the MCU providing the conference audio to the plurality of videoconferencing endpoints.

A multi-point control unit (MCU), comprising:

• a processor;
• one or more inputs coupled to the processor;
• one or more decoders coupled to the processor;
• one or more outputs coupled to the processor;
• one or more encoders coupled to the processor;
• and
• a memory medium coupled to the processor, wherein the memory medium stores program instructions for performing a videoconference, wherein the program instructions are executable by the processor to implement:
• hosting a conference between a plurality of participants at respective participant locations, wherein the conference is performed using a plurality of endpoints at the respective participant locations;
• receiving encoded audio from each of the plurality of endpoints via the one or more inputs;
• receiving audio level information from each of the plurality of endpoints;
• decoding the encoded audio from a plural subset of the plurality of endpoints using the one or more decoders to produce decoded audio, wherein the audio from the plural subset is decoded based on the audio level information, and wherein the plural subset is less than all of the plurality of endpoints, said decoding comprising analyzing the audio level information from each of the plurality of endpoints to determine audio from a designated number (n) of videoconferencing endpoints which have highest audio levels, wherein the value of n is determined dynamically based on said analyzing, and wherein said combining the audio from the plural subset of the plurality of endpoints comprises combining only the audio from the n endpoints;
• encoding the decoded audio using the one or more encoders to produce conference audio;
• and
• providing the conference audio to the plurality of endpoints via the one or more outputs.