초록 ▼

Devices, software and methods are provided for generating aggregate comfort noise for teleconferencing over IP networks. A transcoding component includes a decoder for decoding streams of packets. A summing component has a summer with summing inputs to receive the decoded streams of packets. The su

Devices, software and methods are provided for generating aggregate comfort noise for teleconferencing over IP networks. A transcoding component includes a decoder for decoding streams of packets. A summing component has a summer with summing inputs to receive the decoded streams of packets. The summing component has at least one silence flag input, and an additional signaling path is used by the transcoding component to signal to the silence flag input if any of the decoded streams of packets includes a silence identification packet. In another embodiment, the summing component may or may not include the silence flag input, but the device includes an aggregate comfort noise generation component. The aggregate comfort noise may be programmed to be a balanced representation of all background noises.

대표청구항 ▼

The invention claimed is: 1. A teleconferencing device comprising: a transcoding component having a decoder for decoding streams of packets received concurrently from different users via a network; a summing component having a summer with summing inputs to receive the decoded streams of packets an

The invention claimed is: 1. A teleconferencing device comprising: a transcoding component having a decoder for decoding streams of packets received concurrently from different users via a network; a summing component having a summer with summing inputs to receive the decoded streams of packets and at least one silence flag input; and an additional signaling path for the transcoding component to signal to the silence flag input of the summing component if any of the decoded streams of packets includes a silence identification packet. 2. A device comprising: a transcoding component having a decoder for decoding streams of packets received concurrently from different users via a network; a summing component having a summer to combine the decoded streams of packets; a comfort noise generation component for generating an aggregate background noise for at least one of the users; and an additional signaling path for the transcoding component to signal to the comfort noise generation component if any of the decoded streams of packets includes a silence identification packet. 3. The device of claim 2, wherein the comfort noise generation component includes a parameter generation component to generate constituent components for the aggregate background noise from information in the decoded streams of packets. 4. The device of claim 3, wherein the comfort noise generation component further includes a synthesizer to synthesize the constituent components into comfort noise. 5. A teleconferencing device comprising: means for decoding streams of packets received concurrently from different users via a network; means for summing to combine the decoded streams of packets; and means for generating an aggregate background noise for at least one of the users packets, the means for generating including a means for receiving a silence identification packet. 6. The teleconferencing device of claim 5, wherein the means for generating the aggregate background noise includes a means for generating constituent components for the aggregate background noise from information in the decoded streams of packets. 7. The teleconferencing device of claim 6, wherein the means for generating the aggregate background noise further includes a synthesizer to synthesize the constituent components into comfort noise. 8. A device comprising: a network interface for coupling to a network; and a processor coupled with the network interface, wherein the processor is configured to: establish a plurality of concurrent VoIP connections with respective endpoints through the network; receive respective streams of packets from at least two of the endpoints; decode the received streams of packets; combine the decoded streams of packets in a summer; determine whether one of the streams includes a silent voice; and if so, issue a signal to the summer. 9. The device of claim 8, wherein determining is by detecting a silence identification packet in the stream. 10. An article comprising: a storage medium, said storage medium having stored thereon instructions, that, when executed by at least one device, result in: establishing a plurality of concurrent VoIP connections with respective endpoints; receiving respective streams of packets from at least two of the endpoints via their respective VoIP connections; identifying at least one of the received packets as a silence identification packet; indicating from a transcoder to a comfort noise generator when at least one of the received packets is a silence identification packet; and generating constituent parameters of a balanced background noise in response to the identified packet. 11. The article of claim 10, wherein the constituent parameters are generated from source noise parameters of the identified packet. 12. The article of claim 11, wherein the constituent parameters are generated by power averaging over frequency components of the source noise parameters. 13. The article of claim 10, wherein the instructions further result in: synthesizing the constituent parameters into an aggregate background noise; and encoding the synthesized aggregate background noise. 14. The article of claim 10, wherein the instructions further result in: encoding the constituent parameters in a conference silence identification packet; and transmitting the conference silence identification packet to at least one of the endpoints. 15. A method comprising: establishing a plurality of concurrent VoIP connections with respective endpoints; receiving respective streams of packets from at least two of the endpoints via their respective VoIP connections; decoding the received streams of packets; combining the decoded streams of packets in a summer; determining whether one of the streams includes a silent voice; and if so, signaling to the summer. 16. The method of claim 15, wherein determining is by detecting a silence identification packet in the streams. 17. A method comprising: establishing a plurality of concurrent VoIP connections with respective endpoints; receiving respective streams of packets from at least two of the endpoints via their respective VoIP connections; identifying at least one of the received packets as a silence identification packet; and upon identifying the silence identification packet, signaling from a transcoder to a comfort noise generator; and generating constituent parameters of a balanced background noise in response to the identified packet. 18. The method of claim 17, wherein the constituent parameters are generated from source noise parameters of the identified packet. 19. The method of claim 18, wherein the constituent parameters are generated by power averaging over frequency components of the source noise parameters. 20. The method of claim 17, further comprising: synthesizing the constituent parameters into an aggregate background noise; and encoding the synthesized aggregate background noise. 21. The method of claim 17, further comprising: encoding the constituent parameters in a conference silence identification packet; and transmitting the conference silence identification packet to at least one of the endpoints. 22. The method of claim 21, wherein transmitting the conference silence identification packet further includes transmitting the conference silence identification from a teleconferencing device to a plurality of endpoints.