INITIATING ACTIONS BASED ON PARTIAL HOTWORDS

• US 20150154954A1
• Filed: 03/20/2014
• Published: 06/04/2015
• Est. Priority Date: 12/04/2013
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Sharifi, Matthew
• Granted Patent: US 9,330,663 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/28 Constructional details of speech recognition systems
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 25/51 for comparison or discrimination

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, receiving audio data; determining that an initial portion of the audio data corresponds to an initial portion of a hotword; in response to determining that the initial portion of the audio data corresponds to the initial portion of the hotword, selecting, from among a set of one or more actions that are performed when the entire hotword is detected, a subset of the one or more actions; and causing one or more actions of the subset to be performed.

Claims

1-20. -20. (canceled)

21. A computer-implemented method comprising:

receiving audio data;

determining that an initial portion of the audio data corresponds to a partial utterance of a hotword;

in response to determining that the initial portion of the audio data corresponds to the partial utterance of the hotword, selecting a set of one or more actions that are associated with the partial utterance of the hotword;

causing the one or more actions of the set to be performed;

after determining that the initial portion of the audio data corresponds to the partial utterance of the hotword, determining that the audio data does not correspond to an utterance of the hotword;

in response to determining that the audio data does not correspond to the utterance of the hotword, selecting an additional set of one or more actions that counteract the one more actions of the set; and

causing the one or more actions of the additional set to be performed.

22. The computer-implemented method of claim 21, wherein determining that the audio data does not correspond to the utterance of the hotword comprises:

identifying one or more acoustic features of the audio data;

comparing the one or more acoustic features of the audio data to one or more acoustic features associated with one or more hotwords stored in a database; and

based on the comparing, determining that the audio data does not correspond to any hotword of the one or more hotwords stored in the database.

23. The computer-implemented method of claim 21, wherein determining that the audio data does not correspond to the utterance of the hotword comprises:

obtaining a transcription of the audio data;

comparing the transcription of the audio data to one or more transcriptions associated with one or more hotwords stored in a database; and

based on the comparing, determining that the audio data does not correspond to any hotword of the one or more hotwords stored in the database.

24. The computer-implemented method of claim 21, wherein determining that the audio data does not correspond to the utterance of the hotword comprises determining that the audio data does not correspond to a full utterance of the hotword.

25. The computer-implemented method of claim 21, wherein causing the one or more actions of the additional set to be performed comprises negating the actions of the set of the one or more actions.

26. The computer-implemented method of claim 21, wherein a particular action of the set of actions includes initializing a network connection of a computing device, and a corresponding action of the additional set of actions includes closing the network connection of the computing device.

27. The computer-implemented method of claim 21, wherein the one or more actions of the set include actions whose respective results are not discernible to a user of a computing device performing the actions.

28. A computer-implemented method comprising:

identifying a set of one or more actions that are associated with an utterance of a hotword;

identifying a subset of the one or more actions that are to be associated with a partial utterance of the hotword, wherein the subset of actions include one or more actions whose respective results are not discernible to a user of a computing device that performs the actions; and

associating the subset of the actions with the partial utterance of the hotword.

29. The computer-implemented method of claim 28, wherein the subset of actions includes actions that are performed in a background computing environment of the computing device.

30. The computer-implemented method of claim 28, wherein the subset of actions include one or more of initializing a camera functionality of the computing device, activating the computing device from a sleep state, initializing a network connection on the computer device, and preloading an application on the computing device associated with the hotword.

31. The computer-implemented method of claim 28, further comprising:

identifying an additional subset of the one or more actions that are to be associated with a further utterance of the hotword that is subsequent to the partial utterance of the hotword; and

associating the additional subset of the actions with the further utterance of the hotword.

32. The computer-implemented method of claim 31, wherein the set of the one or more actions includes the subset of the one or more actions and the additional subset of the one more actions.

33. The computer-implemented method of claim 31, wherein the additional subset of the one or more actions include one or more actions whose respective results are discernible to the user of the computer device.

34. A computer-implemented method comprising:

receiving audio data;

determining that an initial portion of the audio data corresponds to a partial utterance of a hotword;

in response to determining that the initial portion of the audio data corresponds to the partial utterance of the hotword, selecting a set of one or more actions that are associated with the partial utterance of the hotword;

causing the one or more actions of the set to be performed;

after determining that the initial portion of the audio data corresponds to the partial utterance of the hotword, determining that the audio data corresponds to an utterance of the hotword;

in response to determining that the audio data corresponds to the utterance of the hotword, selecting an additional set of one or more actions that are associated with the utterance of the hotword; and

causing the one or more actions of the additional set to be performed.

35. The computer-implemented method of claim 34, wherein determining that the audio data correspond to the utterance of the hotword comprises:

identifying one or more acoustic features of the audio data;

comparing the one or more acoustic features of the audio data to one or more acoustic features associated with one or more hotwords stored in a database; and

based on the comparing, determining that the audio data corresponds to the hotword of the one or more hotwords stored in the database.

36. The computer-implemented method of claim 34, wherein determining that the audio data corresponds to the utterance of the hotword comprises:

obtaining a transcription of the audio data;

comparing the transcription of the audio data to one or more transcriptions associated with one or more hotwords stored in a database; and

based on the comparing, determining that the audio data corresponds to the hotword of the one or more hotwords stored in the database.

37. The computer-implemented method of claim 34, wherein determining that the audio data corresponds to the utterance of the hotword comprises determining that the audio data corresponds to a full utterance of the hotword.

38. The computer-implemented method of claim 34, wherein the one or more actions of the set include actions whose respective results are not discernible to a user of a computing device performing the actions.

39. The computer-implemented method of claim 34, wherein the hotword is a particular term that triggers semantic interpretation of an additional term of one or more terms that follow the particular term.