Voice commands for transitioning between device states

• US 9,047,857 B1
• Filed: 12/19/2012
• Issued: 06/02/2015
• Est. Priority Date: 12/19/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Rawles, LLC (Amazon.com, Inc.)
• Inventors: Barton, William F.
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US13/720,792
• Time in Patent Office: 895 Days
• Field of Search: 704231-257, 704270-275
• US Class Current: 1/1
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 25/21 the extracted parameters being power information
• CPC Class Codes: G10L 25/48 specially adapted for particular use

Abstract

Techniques for transitioning an electronic device between device states. In one example, a voice-controlled device is configured to transition from a sleep state to an awake state in response to identifying a user speaking a predefined utterance. The techniques may determine whether the user has spoken the predefined utterance with reference to traditional speech-recognition techniques, as well as with reference to changes in the volume of a user'"'"'s voice.

Claims

1. An apparatus comprising:

a microphone unit;

one or more processors; and

one or more computer-readable media storing computer-executable instructions that, when executed by the one or more processors, cause the one or more processors to perform acts comprising;

receiving an audio signal from the microphone unit, the audio signal including a first portion of speech and a second portion of speech subsequent to the first portion, the second portion including a predefined word or phrase;

calculating a volume of the first portion;

calculating a volume of the second portion;

determining a first score indicating a similarity between the second portion of speech and the predefined word or phrase; and

determining a second score based at least in part on the first score, the volume of the first portion, and the volume of the second portion; and

causing the apparatus to transition from a first state to a second state at least partly in response to determining that second score is greater than a threshold.

2. An apparatus as recited in claim 1, wherein:

the first portion comprises a first set of multiple frames;

the volume of the first portion comprises an average volume of the first set of multiple frames;

the second portion comprises a second set of multiple frames; and

the volume of the second portion comprises an average volume of the second set of multiple frames.

3. An apparatus as recited in claim 1, wherein the apparatus performs more functionality in the second state than in the first state.

4. An apparatus as recited in claim 1, wherein the apparatus provides audio signals generated by the microphone unit to one or more remote computing devices when in the second state and does not provide audio signals generated by the microphone unit to the one or more remote computing devices when in the first state.

5. An apparatus as recited in claim 1, wherein the volume of the second portion is calculated at least partly in response to determining that the first score is greater than another threshold.

6. One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:

determining a first score indicating a similarity between (i) an audio signal that includes speech from a user, and (ii) a predefined word or phrase;

determining a first volume of a first portion of the audio signal;

determining a second volume of a second portion of the audio signal;

determining a second score based at least in part on the first score, the first volume, and the second volume; and

instructing an apparatus to perform an operation based at least in part on the second score.

7. One or more non-transitory computer-readable media as recited in claim 6, wherein the second volume corresponds to speech prior to the user speaking the predefined word or phrase.

8. One or more non-transitory computer-readable media as recited in claim 6, wherein determining the second score comprises adjusting the first score based at least in part on the first volume and the second volume.

9. One or more non-transitory computer-readable media as recited in claim 6, wherein the second score is determined based at least in part on a ratio or a difference between the first volume and the second volume.

10. One or more non-transitory computer-readable media as recited in claim 6, wherein the second score is determined using a neural network or a support vector machine (SVM).

11. One or more non-transitory computer-readable media as recited in claim 6, the acts further comprising determining whether a ratio of the first volume to the second volume is greater than a threshold, and wherein the instructing the apparatus to perform the apparatus is also based at least in part on determining that the ratio is greater than the threshold.

12. One or more non-transitory computer-readable media as recited in claim 6, wherein instructing the apparatus to perform the operation comprises instructing the apparatus to transition from a first state to a second state.

13. One or more non-transitory computer-readable media as recited in claim 12, wherein the apparatus provides audio signals to one or more remote computing resources when in the second state, but not when in the first state.

14. One or more non-transitory computer-readable media as recited in claim 12, wherein the apparatus performs speech recognition when operating in the second state.

15. One or more non-transitory computer-readable media as recited in claim 6, wherein:

determining the first volume comprises determining an average volume of a first set of multiple frames of the audio signal; and

determining the second volume comprises determining an average volume of a second set of multiple frames of the audio signal.

16. One or more non-transitory computer-readable media as recited claim 15, wherein determining the first volume comprises excluding frames of the first set of multiple frames that do not include speech.

17. One or more non-transitory computer-readable media as recited in claim 6, wherein the first score is determined at least partly using a Hidden Markov Model (HMM) and a Gaussian mixture model.

18. A method comprising:

under control of one or more computing devices that include executable instructions,

analyzing a first portion of an audio signal to determine a first score indicating a similarity between the first portion of the audio signal and a predefined word or phrase;

determining a volume of the first portion of the audio signal and a volume of a second portion of the audio signal; and

calculating a second score based at least in part on the first score, the volume of the first portion, and the volume of the second portion.

19. A method as recited in claim 18, further comprising:

comparing the second score to a threshold; and

instructing an apparatus to perform an operation at least partly in response to determining that the second score is greater than the threshold.

20. A method as recited in claim 18, further comprising:

comparing the second score a threshold; and

causing the apparatus to transition from a first state to a second state at least partly in response to determining that the second score is greater than the second threshold.

21. A method as recited in claim 20, wherein the one or more computing devices provide audio signals to one or more remote computing resources when in the second state, but not when in the first state.

22. A method as recited in claim 18, wherein:

the one or more computing devices comprise an apparatus within an environment in which the speech occurs;

or

the one or more computing devices form a portion of a network-accessible computing platform that is remote from the environment in which the speech occurs.

23. A method as recited in claim 18, wherein the first score is determined at least partly using a Hidden Markov Model (HMM) and a Gaussian mixture model.

24. A method as recited in claim 18, wherein calculating the second score comprises adjusting the first score based at least in part on the volume of the first portion and the volume of the second portion.

25. A method as recited in claim 18, wherein the second score is calculated based at least in part on a ratio or a difference between the volume of the first portion and the volume of the second portion.

26. A method as recited in claim 18, wherein the second score is calculated using a neural network or a support vector machine (SVM).