Optimization of network microphone devices using noise classification
Systems and methods for optimizing network microphone devices using noise classification are disclosed herein. In one example, individual microphones of a network microphone device (NMD) detect sound. The sound data is analyzed to detect a trigger event such as a wake word. Metadata associated with the sound data is captured in a lookback buffer of the NMD. After detecting the trigger event, the metadata is analyzed to classify noise in the sound data. Based on the classified noise, at least one performance parameter of the NMD is modified.
1. A playback device, comprising:
one or more processors;
one or more microphones; and
a tangible, non-transitory, computer-readable medium storing instructions executable by the one or more processors to cause the playback device to perform operations comprising:
detecting sound via the one or more microphones;
capturing sound data based on the detected sound;
analyzing the sound data to detect a trigger event;
transmitting the sound data to a first one or more remote computing devices associated with a voice assistant service (VAS);
capturing metadata associated with the sound data, wherein the sound data is not derivable from the metadata;
transmitting the metadata absent the sound data to a second one or more remote computing devices associated with a remote evaluator, the remote evaluator being distinct from the VAS;
receiving, from the remote evaluator, results of analysis of the metadata to classify noise in the sound data; and
based on the classified noise, modifying at least one performance parameter of the playback device.
2. The playback device of claim 1 , wherein the results of analysis of the metadata to classify noise in the sound data comprise an identification of a noise source.
3. The playback device of claim 1 , wherein the metadata comprises frequency-domain information derived from the detected sound.
4. The playback device of claim 1 , wherein the metadata comprises a frequency response spectrum.
5. The playback device of claim 1 , wherein the metadata comprises at least one of:
microphone frequency response data;
microphone spectral data;
acoustic echo cancellation (AEC) data;
echo return loss enhancement (ERLE) data;
arbitration data;
signal level data;
or direction detection data.
6. The playback device of claim 1 , wherein modifying the at least one performance parameter of the playback device comprises at least one of:
adjusting a wake-word-detection sensitivity parameter of the playback device;
adjusting a playback volume of a playback device associated with the playback device; or
modifying a noise-cancellation algorithm of the playback device.
7. The playback device of claim 1 , wherein the operations further comprise transmitting data corresponding to the classified noise to one or more remote computing devices over a wide area network.
8. A method comprising:
detecting sound via one or more microphones of a playback device;
capturing sound data via the playback device based on the detected sound;
analyzing, via the playback device, the sound data to detect a trigger event;
transmitting, via the playback device, the sound data to a first one or more remote computing devices associated with a voice assistant service (VAS);
capturing metadata associated with the sound data via the playback device, wherein the sound data is not derivable from the metadata;
transmitting, via the playback device, the metadata absent the sound data to a second one or more remote computing devices associated with a remote evaluator, the remote evaluator being distinct from the VAS;
receiving, at the playback device and from the remote evaluator, results of analysis of the metadata to classify noise in the sound data; and
based on the classified noise, modifying at least one performance parameter of the playback device.
9. The method of claim 8 , wherein the results of analysis of the metadata to classify noise in the sound data comprise an identification of a noise source.
10. The method of claim 8 , wherein the metadata comprises frequency-domain information derived from the detected sound.
11. The method of claim 8 , wherein the metadata comprises a frequency response spectrum.
12. The method of claim 8 , wherein the metadata comprises at least one of:
microphone frequency response data;
microphone spectral data;
acoustic echo cancellation (AEC) data;
echo return loss enhancement (ERLE) data;
arbitration data;
signal level data;
or direction detection data.
13. The method of claim 8 , wherein modifying the at least one performance parameter of the playback device comprises at least one of:
adjusting a wake-word-detection sensitivity parameter of the playback device;
adjusting a playback volume of a playback device associated with the playback device; or
modifying a noise-cancellation algorithm of the playback device.
14. The method of claim 8 , further comprising transmitting, via the playback device, data corresponding to the classified noise to one or more remote computing devices over a wide area network.
15. Tangible, non-transitory, computer-readable medium storing instructions executable by one or more processors to cause a playback device to perform operations comprising:
detecting sound via one or more microphones of the playback device;
capturing sound data via the playback device based on the detected sound;
analyzing, via the playback device, the sound data to detect a trigger event;
transmitting, via the playback device, the sound data to a first one or more remote computing devices associated with a voice assistant service (VAS);
capturing metadata associated with the sound data via the playback device, wherein the sound data is not derivable from the metadata;
transmitting, via the playback device, the metadata absent the sound data to a second one or more remote computing devices associated with a remote evaluator, the remote evaluator being distinct from the VAS;
receiving, at the playback device and from the remote evaluator, results of analysis of the metadata to classify noise in the sound data; and
based on the classified noise, modifying at least one performance parameter of the playback device.
16. The computer-readable medium of claim 15 , wherein the results of analysis of the metadata to classify noise in the sound data comprise an identification of a noise source.
17. The computer-readable medium of claim 15 , wherein the metadata comprises frequency-domain information derived from the detected sound.
18. The computer-readable medium of claim 15 , wherein the metadata comprises at least one of:
microphone frequency response data;
microphone spectral data;
acoustic echo cancellation (AEC) data;
echo return loss enhancement (ERLE) data;
arbitration data;
signal level data;
or direction detection data.
19. The computer-readable medium of claim 15 , wherein modifying the at least one performance parameter of the playback device comprises at least one of:
adjusting a wake-word-detection sensitivity parameter of the playback device;
adjusting a playback volume of a playback device associated with the playback device; or
modifying a noise-cancellation algorithm of the playback device.
20. The computer-readable medium of claim 15 , wherein the operations further comprise transmitting, via the playback device, data corresponding to the classified noise to one or more remote computing devices over a wide area network.