Hotword detection on multiple devices
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for hotword detection on multiple devices are disclosed. In one aspect, a method includes the actions of receiving, by a first computing device, audio data that corresponds to an utterance. The actions further include determining a first value corresponding to a likelihood that the utterance includes a hotword. The actions further include receiving a second value corresponding to a likelihood that the utterance includes the hotword, the second value being determined by a second computing device. The actions further include comparing the first value and the second value. The actions further include based on comparing the first value to the second value, initiating speech recognition processing on the audio data.
1. A computer-implemented method executed on data processing hardware of a first computing device that causes the data processing hardware to perform operations comprising:
receiving audio data that corresponds to an utterance of a voice command and a hotword preceding the voice command, the utterance of the voice command and the hotword preceding the voice command captured by the first computing device and a second computing device, the first computing device and the second computing device each configured to respond to voice commands that are preceded by the hotword;
transmitting, from the first computing device to the second computing device after the second computing device captured the utterance of the hotword preceding the voice command, a message, the message received by the second computing device causing the second computing device to not respond to the voice command despite the second computing device capturing the utterance of the voice command and the hotword preceding the voice command; and
based on the message, causing the first computing device to respond to the voice command.
2. The computer-implemented method of claim 1 , wherein the message transmitted to the second computing device indicates an audio metric related to the utterance of the hotword captured by the first computing device.
3. The computer-implemented method of claim 2 , wherein the audio metric comprises a loudness of the utterance captured by the first computing device.
4. The computer-implemented method of claim 1 , wherein the operations further comprise, after receiving the audio data that corresponds to the utterance of the voice command and the hotword preceding the voice command:
receiving, from the second computing device, an additional message,
wherein causing the first computing device to respond to the voice command is further based on the additional message received from the second computing device.
5. The computer-implemented method of claim 4 , wherein the additional audio metric comprises a loudness of the utterance captured by the first computing device.
6. The computer-implemented method of claim 1 , wherein the operations further comprise, in response to receiving the audio data that corresponds to the utterance of the voice command and the hotword preceding the voice command, determining that the audio data likely includes the utterance of the hotword preceding the voice command without performing automated speech recognition on the audio data.
7. The computer-implemented method of claim 6 , wherein determining that the audio data likely includes the utterance of the hotword preceding the voice command comprises:
determining a hotword score that reflects a likelihood that the audio data includes the utterance of the hotword preceding the voice command; and
determining that the hotword score satisfies a threshold.
8. The computer-implemented method of claim 1 , wherein receiving the audio data comprises receiving the audio data while the first computing device is in a low power mode.
9. The computer-implemented method of claim 1 , wherein the first computing device and the second computing device are in communication via a local network.
10. The computer-implemented method of claim 1 , wherein the operations further comprise determining to perform speech recognition on the audio data.
11. A first computing device comprising:
data processing hardware; and
memory hardware in communication with the data processing hardware and storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:
receiving audio data that corresponds to an utterance of a voice command and a hotword preceding the voice command, the utterance of the voice command and the hotword preceding the voice command captured by the first computing device and a second computing device, the first computing device and the second computing device each configured to respond to voice commands that are preceded by the hotword;
transmitting, from the first computing device to the second computing device after the second computing device captured the utterance of the hotword preceding the voice command, a message, the message received by the second computing device causing the second computing device to not respond to the voice command despite the second computing device capturing the utterance of the voice command and the hotword preceding the voice command; and
based on the message, causing the first computing device to respond to the voice command.
12. The first computing device of claim 11 , wherein the message transmitted to the second computing device indicates an audio metric related to the utterance of the hotword captured by the first computing device.
13. The first computing device of claim 12 , wherein the audio metric comprises a loudness of the utterance captured by the first computing device.
14. The first computing device of claim 11 , wherein the operations further comprise, after receiving the audio data that corresponds to the utterance of the voice command and the hotword preceding the voice command:
receiving, from the second computing device, an additional message,
wherein causing the first computing device to respond to the voice command is further based on the additional message received from the second computing device.
15. The first computing device of claim 14 , wherein the additional audio metric comprises a loudness of the utterance captured by the first computing device.
16. The first computing device of claim 11 , wherein the operations further comprise, in response to receiving the audio data that corresponds to the utterance of the voice command and the hotword preceding the voice command, determining that the audio data likely includes the utterance of the hotword preceding the voice command without performing automated speech recognition on the audio data.
17. The first computing device of claim 16 , wherein determining that the audio data likely includes the utterance of the hotword preceding the voice command comprises:
determining a hotword score that reflects a likelihood that the audio data includes the utterance of the hotword preceding the voice command; and
determining that the hotword score satisfies a threshold.
18. The first computing device of claim 11 , wherein receiving the audio data comprises receiving the audio data while the first computing device is in a low power mode.
19. The first computing device of claim 11 , wherein the first computing device and the second computing device are in communication via a local network.
20. The first computing device of claim 11 , wherein the operations further comprise determining to perform speech recognition on the audio data.