US 10,573,321 B1Granted

Voice detection optimization based on selected voice assistant service

AssigneeSonos, Inc.

InventorsConnor Kristopher Smith, Kurt Thomas Soto, Charles Conor Sleith

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods for optimizing voice detection via a network microphone device (NMD) based on a selected voice-assistant service (VAS) are disclosed herein. In one example, the NMD detects sound via individual microphones and selects a first VAS to communicate with the NMD. The NMD produces a first sound-data stream based on the detected sound using a spatial processor in a first configuration. Once the NMD determines that a second VAS is to be selected over the first VAS, the spatial processor assumes a second configuration for producing a second sound-data stream based on the detected sound. The second sound-data stream is then transmitted to one or more remote computing devices associated with the second VAS.

Claims20

Dependent claims

A playback device comprising:

a plurality of microphones;
a network interface;
one or more processors;
and
tangible, non-transitory computer-readable media having stored therein instructions executable by the one or more processors to cause the playback device to perform a method comprising:
capturing audio via a first set of microphones selected from the plurality of microphones;
analyzing the audio captured via the first set of microphones using a first wake-word engine on the playback device to detect a first wake word;
selecting a second wake-word engine on the playback device, wherein the second wake-word engine is different from the first wake-word engine;
after selecting the second wake-word engine, capturing audio via a second set of microphones selected from the plurality of microphones, wherein the second set of microphones is different from the first set of microphones;
analyzing the audio captured via the second set of microphones using the second wake-word engine to detect a second wake word;
detecting a wake word via one of the first wake-word engine or the second wake-word engine, wherein the detected wake word comprises one of the first wake word or the second wake word;
and
transmitting, via the network interface, at least a voice utterance following the detected wake word to one or more remote servers corresponding to a particular voice assistant service associated with the detected wake word.

10 more elements

A tangible, non-transitory computer-readable medium having stored therein instructions executable by one or more processors to cause a playback device to perform a method comprising:

capturing audio via a first set of microphones selected from a plurality of microphones of the playback device;
analyzing the audio captured via the first set of microphones using a first wake-word engine on the playback device to detect a first wake word;
selecting a second wake-word engine on the playback device, wherein the second wake-word engine is different from the first wake-word engine;
after selecting the second wake-word engine, capturing audio via a second set of microphones selected from the plurality of microphones, wherein the second set of microphones is different from the first set of microphones;
analyzing the audio captured via the second set of microphones using the second wake-word engine to detect a second wake word;
detecting a wake word via one of the first wake-word engine or the second wake-word engine, wherein the detected wake word comprises one of the first wake word or the second wake word;
and
transmitting, via a network interface of the playback device, at least a voice utterance following the detected wake word to one or more remote servers corresponding to a particular voice assistant service associated with the detected wake word.

5 more elements

A method comprising:

capturing audio via a first set of microphones selected from a plurality of microphones of a playback device;
analyzing the audio captured via the first set of microphones using a first wake-word engine on the playback device to detect a first wake word;
selecting a second wake-word engine on the playback device, wherein the second wake-word engine is different from the first wake-word engine;
after selecting the second wake-word engine, capturing audio via a second set of microphones selected from the plurality of microphones, wherein the second set of microphones is different from the first set of microphones;
analyzing the audio captured via the second set of microphones using the second wake-word engine to detect a second wake word;
detecting a wake word via one of the first wake-word engine or the second wake-word engine, wherein the detected wake word comprises one of the first wake word or the second wake word;
and
transmitting, via a network interface of the playback device, at least a voice utterance following the detected wake word to one or more remote servers corresponding to a particular voice assistant service associated with the detected wake word.

5 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/434,426Filed 7 Jun 2019 · continuation
Priority: 25 Sep 2018
Granted: 25 Feb 2020
Term ends: Not estimatedThe source records don’t support an estimate.

Classification19 codes

G10L 15/30
Constructional details of speech recognition systems: distributed recognition, e.g. in client-server systems, for mobile phones or network…
G10L 15/08
Speech recognition: speech classification or search
G10L 15/22
Speech recognition: procedures used during a speech recognition process, e.g. man-machine dialogue
H04R 1/406
By combining a number of identical transducers: microphones