Background noise reduction in voice communication

1. A system configured to cancel background noise in voice communication, the system comprising:

a user device configured to capture a background sound in the voice communication; and

a server configured to:

receive the background sound captured by the user device;

identify a source of the background sound from among a plurality of sound sources;

generate a reference background sound by adjustment of a sound corresponding to the identified source based on a quality of sound for the background sound and an impulse response of the identified source, wherein the impulse response represents channel information of a channel from the source of the background sound to a sound detector of the user device; and

cancel the background sound in the voice communication based at least in part on the generated reference background sound.

2. The system of claim 1 , wherein the user device is further configured to cancel the background sound in the voice communication based at least in part on the generated reference background sound.

3. The system of claim 1 , further comprising:

another user device configured to participate in the voice communication and cancel the background sound in the voice communication based at least in part on the generated reference background sound.

4. The system of claim 1 , wherein the user device is configured to capture the background sound in an absence of a voice signal from a user of the user device, after a voice call request signal is received as part of an incoming call or after the user device enters a call mode as part of an outgoing call.

5. The system of claim 1 , wherein the server is configured to identify the source of the background sound by comparison of the background sound to sounds corresponding to the plurality of sound sources.

6. The system of claim 5 , wherein the server is configured to compare the background sound to the sounds corresponding to the plurality of sound sources by comparison of an audio fingerprint of the background sound to audio fingerprints of the sounds corresponding to the plurality of sound sources.

7. The system of claim 1 , wherein the server is further configured to estimate properties of the captured background sound by use of a channel estimation technique.

8. The system of claim 1 , wherein the user device is further configured to cancel the background sound based at least in part on the generated reference background sound by use of an adaptive noise cancellation technique.

9. A server, comprising:

a receiver configured to receive a background sound from a user device;

a sound retrieval unit configured to identify a source of the background sound from among a plurality of sound sources by comparison of the background sound to sounds corresponding to the plurality of sound sources;

a reference sound generation unit configured to generate a reference background sound by adjustment of a sound corresponding to the identified source based on a quality of sound for the background sound and an impulse response of the identified source, wherein the impulse response represents channel information of a channel from the source of the background sound to a sound detector of the user device;

a transmitter configured to transmit the reference background sound to the user device; and

a noise canceller configured to cancel the background sound in a voice communication based at least in part on the generated reference background sound.

10. The server of claim 9 , wherein the sound retrieval unit is configured to compare the background sound to the sounds corresponding to the plurality of sound sources by comparison of an audio fingerprint of the background sound to audio fingerprints of the sounds corresponding to the plurality of sound sources.

11. The server of claim 9 , further comprising, a property estimation unit configured to estimate properties of the background sound by use of a channel estimation technique.

12. A server method to cancel background noise in voice communication, the method comprising:

receiving, by a server from a user device, a background sound in the voice communication;

identifying, by the server, a source of the received background sound from among a plurality of sound sources;

generating, by the server, a reference background sound by adjusting a sound corresponding to the identified source based on a quality of sound and an impulse response of the identified sound source, wherein the impulse response represents channel information of a channel from the source of the background sound to a sound detector of the user device; and

cancelling, by the server, the background sound in the voice communication based at least in part on the generated reference background sound.

13. The method of claim 12 , wherein the receiving the background sound includes receiving the background sound captured by the user device in an absence of a voice signal from a user of the user device, after a voice call request signal is received as part of an incoming call or after the user device enters a call mode as part of an outgoing call.

14. The method of claim 12 , wherein the identifying the source of the received background sound includes comparing the received background sound to sounds corresponding to the plurality of sound sources.

15. The method of claim 14 , wherein the comparing the received background sound to the sounds corresponding to the plurality of sound sources includes comparing an audio fingerprint of the received background sound to audio fingerprints of the sounds corresponding to the plurality of sound sources.

16. The method of claim 12 , wherein the generating the reference background sound includes estimating the properties of the received background sound using a channel estimation technique.

17. The method of claim 12 , wherein the cancelling the background sound includes cancelling the background sound based at least in part on the reference background sound using an adaptive noise cancellation technique.

18. A server-based non-transitory computer-readable storage medium which stores a program to cause a processor to cancel or control cancellation of background noise for voice communication, the program comprising one or more instructions to perform operations that comprise:

identifying, by a server, a background sound received from a user device and present in the voice communication;

identifying, by the server, a source of the identified background sound from among a plurality of sound sources;

generating, by the server, a reference background sound by adjusting a sound corresponding to the identified source based on a quality of sound for the background sound and an impulse response of the identified source, wherein the impulse response represents channel information of a channel from the source of the background sound to a sound detector of the user device; and

cancelling, by the server, the background sound in the voice communication based at least in part on the reference background sound.