Altering audio based on non-speech commands

• US 9,087,520 B1
• Filed: 12/13/2012
• Issued: 07/21/2015
• Est. Priority Date: 12/13/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Rawles, LLC (Amazon.com, Inc.)
• Inventors: Salvador, Stan Weidner
• Primary Examiner(s): McFadden, Susan
• Application Number:
  US13/714,236
• Time in Patent Office: 950 Days
• Field of Search: 704/275
• US Class Current: 1/1
• CPC Class Codes: G10L 25/84 for discriminating voice from noise

Abstract

Techniques for altering audio being output by an audio-controlled device, or another device, to enable more accurate automatic speech recognition (ASR) by the audio-controlled device. For instance, an audio-controlled device may output audio within an environment using a speaker of the device. While outputting the audio, a microphone of the device may capture sound within the environment and may generate an audio signal based on the captured sound. The device may then analyze the audio signal to identify a predefined non-speech command issued by a user within the environment. In response to identifying the predefined non-speech command, the device may somehow alter the output of the audio for the purpose of reducing the amount of noise within subsequently captured sound.

Claims

1. An apparatus comprising:

a speaker to output audio in an environment;

a microphone unit to capture sound in the environment, the sound including the audio being output by the speaker and a clapping sound issued by a user in the environment;

a processor; and

computer-readable media storing computer-executable instructions that, when executed by the processor, cause the processor to perform acts comprising;

receiving an audio signal generated by the microphone unit;

identifying, from the audio signal, the clapping sound issued by the user;

attenuating the audio being output by the speaker at least partly in response to the identifying; and

identifying, from the audio signal or from the additional audio signal and while the audio being output by the speaker is attenuated, one or more predefined words spoken by the user.

2. An apparatus as recited in claim 1, the acts further comprising:

providing a subsequent audio signal to one or more computing devices that are remote from the environment at least partly in response to the identifying of the one or more predefined words, the one or more computing devices to perform speech recognition on the subsequent audio signal.

3. An apparatus as recited in claim 1, wherein the clapping sound comprises the user clapping the user'"'"'s hands a predefined number of times.

4. An apparatus as recited in claim 1, wherein the clapping sound comprises the user clapping the user'"'"'s hands in a predefined pattern.

5. An apparatus comprising:

a speaker to output audio in an environment;

a microphone unit to capture sound in the environment, the sound including a user in the environment issuing a non-speech command;

a processor; and

computer-readable media storing computer-executable instructions that, when executed by the processor, cause the processor to perform acts comprising;

receiving an audio signal generated by the microphone unit;

identifying, from the audio signal, the non-speech command issued by the user;

altering the output of the audio at least partly in response to the identifying; and

identifying, from the audio signal or from the additional audio signal and while the output of the audio by the speaker is altered, one or more predefined words spoken by the user.

6. An apparatus as recited in claim 5, the acts further comprising:

providing a subsequent audio signal to one or more computing devices that are remote from the environment at least partly in response to the identifying of the one or more predefined words, the one or more computing devices to perform speech recognition on the subsequent audio signal.

7. An apparatus as recited in claim 5, wherein the non-speech command comprises the user clapping the user'"'"'s hands.

8. An apparatus as recited in claim 5, wherein the non-speech command comprises the user whistling.

9. An apparatus as recited in claim 5, wherein the non-speech command comprises the user striking an object in the environment.

10. An apparatus as recited in claim 5, wherein the altering comprises attenuating the audio.

11. An apparatus as recited in claim 5, wherein the altering comprises pausing the audio.

12. An apparatus as recited in claim 5, wherein:

the apparatus comprises two speakers, the two speakers outputting the audio in stereo; and

the altering comprises switching the output of the audio from stereo to mono.

13. An apparatus as recited in claim 5, wherein:

the apparatus comprises two speakers, each of the two speakers outputting at least a portion of the audio; and

the altering comprises turning off at least one speaker.

14. An apparatus as recited in claim 5, the acts further comprising returning the audio to its state prior to the altering, the returning occurring a predefined amount of time after the altering.

15. An apparatus as recited in claim 5, the acts further comprising returning the audio to its state prior to the altering, the returning occurring after a predefined amount of time that does not include speech from the user.

16. An apparatus as recited in claim 5, the acts further comprising:

again identifying the non-speech command from the user after the altering; and

returning the audio to its state prior to the altering at least partly in response to again identifying the non-speech command.

17. A method comprising:

under control of an electronic device that includes a microphone unit, a speaker and executable instructions,

outputting audio via the speaker

sound captured by the determining that a user has issued a non-speech command based at least in part on microphone unit;

altering the output of the audio based at least in part on determining that the user has issued the non-speech command; and

identifying, from the audio signal or from the additional audio signal and while the audio being output via the speaker is altered, one or more predefined words spoken by the user.

18. A method as recited in claim 17, wherein the non-speech command comprises the user clapping in a predefined pattern or striking an object in a predefined pattern.

19. A method as recited in claim 17, wherein the non-speech command comprises the user clapping a predefined number of times or striking an object a predefined number of times.

20. A method as recited in claim 17, wherein the non-speech command comprises the user whistling for a certain amount of time or in a certain frequency.

21. A method as recited in claim 17, wherein the non-speech command comprises the user whistling, the whistling either increasing in frequency over time or decreasing in frequency over time.

22. A method as recited in claim 17, wherein the altering comprises:

switching the output of the audio from stereo to mono;

or

pausing the audio.