Granted Patent B2

US 12,499,901 · App. 18/693,411 · Granted Dec 16, 2025

Inventors: James Nesfield (Edinburgh, GB); Ian Ward Frank (Arlington, MA)

Assignee: Sonos, Inc.

G10L21/0216G10L25/30G10L25/78H04R3/005G10L2021/02165

Abstract

Claims (51)

1 . One or more playback devices comprising:

a first one or more microphones;

a second one or more microphones;

one or more processors; and

data storage having instructions stored thereon that, when executed by the one or more processors, cause the one or more playback devices to perform operations comprising:

capturing a first sound data stream based on a user voice input via the first one or more microphones;

while capturing the first sound data stream, (i) concurrently capturing a second sound data stream based on the user voice input via the second one or more microphones and (ii) evaluating the first sound data stream to determine whether a noise threshold is exceeded;

while the noise threshold is not exceeded, training a synthetic sound data model based on the first sound data stream and the second sound data stream concurrently captured with first sound data;

communicating an output audio stream to at least one second playback device based on the first sound data stream;

detecting noise in the first sound data stream; and

mixing a synthetic audio stream into the output audio stream based on the noise in the first sound data stream, wherein the synthetic audio stream is produced based on the synthetic sound data model.

2 . The one or more playback devices of claim 1 , wherein the operations further comprise:

detecting that the noise is no longer present in the first sound data stream; and

ceasing the mixing of the synthetic audio stream into the output audio stream after detecting that the noise is no longer present in the first sound data stream.

3 . The one or more playback devices of claim 1 , wherein the first one or more microphones comprises an air-conduction microphone and the second one or more microphones comprises a bone-conduction microphone.

4 . The one or more playback devices of claim 1 , wherein the one or more playback devices comprises a portable playback device comprising the first and second microphones.

5 . The one or more playback devices of claim 1 , wherein the mixing is based at least in part on a determined noise level in the second sound data stream.

6 . The one or more playback devices of claim 1 , wherein the mixing comprises utilizing the synthetic sound data model over a first frequency range and utilizing the second sound data stream over a second frequency range different from the first.

7 . The one or more playback devices of claim 1 , wherein the mixing comprises combining the first sound data stream at a first relative volume level with the synthetic audio stream at a second relative volume level.

8 . The one or more playback devices of claim 1 , wherein the mixing comprises dynamically switching between the first sound data stream and the synthetic audio stream.

9 . A method comprising:

capturing a first sound data stream based on a user voice input via a first one or more microphones of at least one first playback device;

while capturing the first sound data stream, (i) concurrently capturing a second sound data stream based on the user voice input via a second one or more microphones of the at least one first playback device and (ii) evaluating the first sound data stream to determine whether a noise threshold is exceeded;

while the noise threshold is not exceeded, training a synthetic sound data model based on the first sound data stream and the second sound data stream concurrently captured with first sound data;

communicating an output audio stream to at least one second playback device based on the first sound data stream;

detecting noise in the first sound data stream; and

mixing a synthetic audio stream into the output audio stream based on the noise in the first sound data stream, wherein the synthetic audio stream is produced based on the synthetic sound data model.

10 . The method of claim 9 , further comprising:

detecting that the noise is no longer present in the first sound data stream; and

ceasing the mixing of the synthetic audio stream into the output audio stream after detecting that the noise is no longer present in the first sound data stream.

11 . The method of claim 9 , wherein the first one or more microphones comprises an air-conduction microphone and the second one or more microphones comprises a bone-conduction microphone.

12 . The method of claim 9 , wherein the at least one first playback device comprises a portable playback device comprising the first and second microphones.

13 . The method of claim 9 , wherein the mixing is based at least in part on a determined noise level in the second sound data stream.

14 . The method of claim 9 , wherein the mixing comprises utilizing the synthetic sound data model over a first frequency range and utilizing the second sound data stream over a second frequency range different from the first.

15 . The method of claim 9 , wherein the mixing comprises combining the first sound data stream at a first relative volume level with the synthetic audio stream at a second relative volume level.

16 . The method of claim 9 , wherein the mixing comprises dynamically switching between the first sound data stream and the synthetic audio stream.

17 . One or more tangible, non-transitory computer-readable media comprising instructions that, when executed by one or more processors of at least one playback device, cause the at least one playback device to perform operations comprising:

capturing a first sound data stream based on a user voice input via a first one or more microphones of the at least one playback device;

while capturing the first sound data stream, (i) concurrently capturing a second sound data stream based on the user voice input via a second one or more microphones of the at least one playback device and (ii) evaluating the first sound data stream to determine whether a noise threshold is exceeded;

while the noise threshold is not exceeded, training a synthetic sound data model based on the first sound data stream and the second sound data stream concurrently captured with first sound data;

communicating an output audio stream to at least one second playback device based on the first sound data stream;

detecting noise in the first sound data stream; and

mixing a synthetic audio stream into the output audio stream based on the noise in the first sound data stream, wherein the synthetic audio stream is produced based on the synthetic sound data model.

18 . The computer-readable media of claim 17 , wherein the operations further comprise:

detecting that the noise is no longer pres 3 nt in the first sound data stream; and

ceasing the mixing of the synthetic audio stream into the output audio stream after detecting that the noise is no longer present in the first sound data stream.

19 . The computer-readable media of claim 17 , wherein the mixing is based at least in part on a determined noise level in the second sound data stream.

20 . The computer-readable media of claim 17 , wherein the mixing comprises at least one of:

utilizing the synthetic sound data model over a first frequency range and utilizing the second sound data stream over a second frequency range different from the first;

combining the first sound data stream at a first relative volume level with the synthetic audio stream at a second relative volume level; or

dynamically switching between the synthetic audio stream and the second sound data stream.

Assignments (2)

Continuity (2)

References Cited (67)