Command suggestions during automatic speech recognition

• US 9,378,740 B1
• Filed: 09/30/2014
• Issued: 06/28/2016
• Est. Priority Date: 09/30/2014
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Rosen, Alexander David, Yin, Yuwang
• Primary Examiner(s): Bryar, Jeremiah
• Application Number:
  US14/502,572
• Time in Patent Office: 637 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/90324 using system suggestions
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Features are disclosed for identifying and providing command suggestions during automatic speech recognition. As utterances are interpreted, suggestions may be provided based on even partial interpretations to guide users of a client device to commands available via speech recognition.

Claims

1. A system comprising:

a computer-readable memory storing executable instructions; and

one or more physical computer processors in communication with the computer-readable memory, wherein the one or more physical computer processors are programmed by the executable instructions to at least;

receive audio data from a client device;

perform automatic speech recognition to obtain a transcript of the audio data, the transcript including first text for the audio data;

select an intent definition from a plurality of intent definitions, the intent definition selected based on a first textual expression included in the intent definition and on the first text;

generate second text using the first text and the intent definition, wherein a portion of the second text matches the first text, wherein the second text corresponds to a second textual expression included in the intent definition; and

transmit the second text to the client device.

2. The system of claim 1, wherein the one or more physical computer processors are further programmed by the executable instructions to generate third text using the first text and the intent definition, wherein a portion of the third text matches the first text, wherein the third text corresponds to a third textual expression included in the intent definition, and wherein the third textual expression includes more information than provided via the first textual expression identified by the first intent definition.

3. The system of claim 1, wherein the one or more physical computer processors are further programmed by the executable instructions to:

select a second intent definition from the plurality of intent definitions, wherein the second intent definition is identified by the first text; and

generate third text using the first text and the second intent definition, wherein a portion of the third text matches the first text, wherein the third text corresponds to one of the second textual expressions included in the second intent definition.

4. The system of claim 1, wherein the first intent definition further includes a slot definition identifying a data type, and

wherein the one or more physical computer processors are further programmed by the executable instructions to:

obtain a user identifier for the client device;

obtain an example text value of the data type for the slot definition based on the user identifier; and

generate the second text using the example text value in the slot.

5. The system of claim 4, wherein the user identifier comprises one of a username, a media access control identifier, a token, or a browser cookie.

6. The system of claim 4, wherein obtaining the example text value of the data type comprises obtaining the example text value from a content library for the identified user.

7. The system of claim 1, wherein the one or more physical computer processors are further programmed by the executable instructions to:

perform automatic speech recognition to obtain a second transcription of the audio data, the second transcription including alternate text for the audio data; and

select the first intent definition using the alternate text.

8. A computer-implemented method comprising:

under control of one or more computing devices configured with specific computer-executable instructions,

performing automatic speech recognition using audio data received from a client device to obtain a transcript of the audio data, the transcript including first text for the audio data;

selecting an intent definition from a plurality of intent definitions, the intent definition selected based on a first textual expression included in the intent definition and on the first text; and

generating second text using the first text and the intent definition, wherein the second text corresponds to a second textual expression included in the intent definition, and wherein a portion of the second text matches the first text.

9. The computer-implemented method of claim 8, wherein the intent definition includes a slot definition identifying a data type, and

wherein the method further comprises:

identifying a user of the client device;

obtaining an example text value of the data type for the slot definition based on the identified user; and

generating the second text comprises using the example text value.

10. The computer-implemented method of claim 9, wherein identifying the user comprises receiving one of a username, a device identifier of the client device, a token, or a browser cookie.

11. The computer-implemented method of claim 9, wherein obtaining the example text value of the data type comprises obtaining the example text value from a content library for the identified user.

12. The computer-implemented method of claim 8, further comprising selecting additional intent definitions based on a comparison of a current resource level for one of the speech processing system or the client device to a predetermined resource level.

13. The computer-implemented method of claim 8, wherein performing automatic speech recognition comprises generating alternate text corresponding to the received audio data, and wherein the intent definition is selected using the alternate text corresponding to the received audio data.

14. The computer-implemented method of claim 8, further comprising generating third text using the first text and the intent definition, wherein a portion of the third text matches the first text, wherein the third text corresponds to a third textual expression included in the intent definition, and wherein the third textual expression includes more information than provided via the first textual expression identified by the first intent definition.

15. The computer-implemented method of claim 8, further comprising:

selecting a second intent definition from the plurality of intent definitions, and wherein the second intent definition is identified by the first text; and

generating third text using the first text and the second intent definition, wherein a portion of the third text matches the first text, wherein the third text corresponds to one of the second textual expressions included in the second intent definition.

16. A non-transitory computer readable medium storing a computer-executable module that, when executed by a processor of a computing device, cause the computing device to perform a process comprising:

performing automatic speech recognition using audio data received from a client device to obtain a transcript of the audio data, the transcript including first text for the audio data;

selecting an intent definition from a plurality of intent definitions, the intent definition selected based on a first textual expression included in the intent definition and on the first text;

generating second text using the first text and the intent definition, wherein the second text corresponds to a second textual expression included in the intent definition, and wherein a portion of the second text matches the first text.

17. The non-transitory computer readable medium of claim 16, wherein the computer-executable module causes the computing device to store the intent definition by storing a slot definition identifying a data type, and

wherein the process further comprises:

identifying a user of the client device;

obtaining an example text value of the data type for the slot definition based on the identified user; and

generating the second text comprises using the example text value.

18. The non-transitory computer readable medium of claim 17, wherein the computer-executable module causes the computing device to obtain the example text value of the data type by obtaining the example text value from a content library for the identified user.

19. The non-transitory computer readable medium of claim 16, wherein the computer-executable module further causes the computing device to perform the process further comprising:

generating third text using the first text and the intent definition, wherein a portion of the third text matches the first text, wherein the third text corresponds to a third textual expression included in the intent definition, and wherein the third textual expression includes more information than provided via the first textual expression identified by the first intent definition.

20. The non-transitory computer readable medium of claim 16, wherein the process further comprises:

selecting a second intent definition from the plurality of intent definitions, and wherein the second intent definition is identified by the first text; and

generating third text using the first text and the second intent definition, wherein a portion of the third text matches the first text, wherein the third text corresponds to one of the second textual expressions included in the second intent definition.

21. The non-transitory computer readable medium of claim 16, wherein the computer-executable module causes the computing device to perform automatic speech recognition by generating alternate text corresponding to the received audio data, and wherein intent definition is selected using the alternate text corresponding to the received audio data.