Keyword detection
Systems, apparatuses, and methods are described for keyword detection. A first detector device may receive audio and process the audio to determine a confidence value regarding presence of a keyword. A copy of the audio may be passed to a second detector device to perform additional testing for the audio. The first detector device may take an action before or after sending to the second detector device based on the confidence value.
1. A method comprising:
determining, by a first detector, a confidence value associated with recognition of a voice command in audio;
sending, to a second detector, a request to process the audio to recognize the voice command; and
prior to receiving a response from the second detector indicating whether the voice command was recognized in the audio, and based on a comparison of the confidence value to a threshold, causing performance of the voice command.
2. The method of claim 1 , further comprising:
reducing an audio level of an audio output;
detecting additional audio after the reducing the audio level of the audio output; and
sending the additional audio to the second detector for processing the additional audio.
3. The method of claim 1 , wherein the voice command comprises a wake keyword, and wherein the method further comprises temporarily lowering an audio volume while one or more additional keywords are spoken following the wake keyword.
4. The method of claim 1 , further comprising:
receiving an indication that the voice command was not recognized by the second detector; and
based on the indication, causing a corrective action to reverse the performance of the voice command.
5. The method of claim 1 , further comprising:
receiving an indication that the voice command was not recognized by the second detector;
determining that it is not too late to prevent performance of the voice command; and
based on the indication and the determining that it is not too late, preventing performance of the voice command.
6. A method comprising:
determining, by a first detector, based on received audio, and based on keyword detection settings of the first detector, a confidence score indicating whether a keyword has been detected;
sending, to a second detector, the confidence score and the audio;
prior to receiving a response from the second detector indicating whether the keyword was recognized in the audio, and based on determining that the confidence score satisfies a threshold, lowering a volume level of an output device for a time period;
receiving, from the second detector, keyword detection information indicating whether the keyword was detected in the audio by the second detector; and
causing, based on the keyword detection information received from the second detector, adjustment of the threshold.
7. The method of claim 6 , further comprising adjusting, based on determining whether the output device is outputting content, the threshold.
8. The method of claim 6 , wherein the keyword detection information indicates that the keyword was not detected in the audio by the second detector; and
based on the indication, shortening the time period.
9. The method of claim 6 , further comprising:
determining, based on the audio, an indication of a change of location of the first detector with respect to the output device; and
deactivating a beam corresponding to a location of the output device.
10. The method of claim 6 , further comprising:
determining, based on the audio, an indication of a location of a user and a location of the output device;
turning off a first beam of the first detector, wherein the first beam corresponds to the location of the output device; and
turning on a second beam of the first detector, wherein the second beam corresponds to the location of the output device.
11. The method of claim 6 , further comprising:
determining, based on the audio, that a user has moved from a first location to a second location;
turning off a first beam, of the first detector, corresponding to the first location; and
turning on a second beam corresponding to the second location.
12. The method of claim 6 , further comprising:
determining, based on the audio, that a rerun is being output on the output device; and
adjusting the threshold to make the threshold more easily satisfied.
13. The method of claim 6 , further comprising:
determining, based on the audio, that a user's preferred content item is being output via the output device; and
based on the determining, making the threshold satisfied less easily.
14. The method of claim 6 , further comprising:
determining that it is not too late to undo the lowering the volume level; and
causing, based on the keyword detection information, the first detector to undo the lowering the volume level.
15. A method comprising:
receiving, by a first detector at a location, audio;
determining, based on the audio and keyword detection settings of the first detector, a confidence score indicating whether a keyword has been detected;
sending, to a second detector, the confidence score and the audio;
based on a comparison of the confidence score to a threshold, maintaining a volume level of a content item being output in the location; and
receiving, from the second detector and after the maintaining the volume level, keyword detection information indicating whether the keyword was confirmed by the second detector.
16. The method of claim 15 , further comprising:
determining, based on the keyword detection information, that the maintaining the volume level of the content item was a mistake; and
based on the determining, lowering the volume level to permit completion of a spoken command.
17. The method of claim 15 , further comprising:
determining, based on the audio, that the content item comprises a rerun; and
based on the determining, adjusting the threshold to make the threshold more easily satisfied.
18. The method of claim 15 , further comprising:
determining, based on the audio, that the content item comprises a user's preferred content item; and
based on the determining, adjusting the threshold to make the threshold less easily satisfied.
19. The method of claim 15 , further comprising:
determining, based on the audio, an indication of a change of location of the first detector with respect to an output device; and
based on the determining, deactivating a beam corresponding to a location of the output device.
20. The method of claim 1 , wherein the threshold indicates whether the first detector should wait for confirmation from the second detector before the performance of the voice command.