Method and apparatus for assigning keyword model to voice operated function

• US 9,786,296 B2
• Filed: 12/10/2013
• Issued: 10/10/2017
• Est. Priority Date: 07/08/2013
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Kim, Taesu, Lee, Minsub
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Le, Thuykhanh
• Application Number:
  US14/101,869
• Publication Number:
  US 20150012279A1
• Time in Patent Office: 1,400 Days
• Field of Search: None
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/225 Feedback of the input speech
• CPC Class Codes: G10L 25/48 specially adapted for particular use

Abstract

A method, performed in an electronic device, for assigning a target keyword to a function is disclosed. In this method, a list of a plurality of target keywords is received at the electronic device via a communication network, and a particular target keyword is selected from the list of target keywords. Further, the method may include receiving a keyword model for the particular target keyword via the communication network. In this method, the particular target keyword is assigned to a function of the electronic device such that the function is performed in response to detecting the particular target keyword based on the keyword model in an input sound received at the electronic device.

Claims

1. A method comprising:

receiving, at an electronic device, a list of a plurality of target keywords, the list received via a communication network;

receiving, at the electronic device, user input that selects a particular target keyword from the list of the plurality of target keywords; and

in response to the user input;

receiving a keyword data file at the electronic device via the communication network, the keyword data file including a keyword model for the particular target keyword and including compatibility information;

determining whether the keyword data file is compatible with a keyword detection unit of the electronic device based on the compatibility information;

receiving a second version of the keyword model based on a determination that the keyword data file is not compatible with the keyword detection unit; and

assigning, at the electronic device, the particular target keyword to a function of the electronic device such that the function is activated in response to detecting, based on the keyword model, that an input sound received at the electronic device includes the particular target keyword.

2. The method of claim 1, further comprising:

receiving response sound data at the electronic device;

detecting whether the particular target keyword is included in the input sound; and

, in response to detecting that the particular target keyword is included in the input sound, outputting a response sound at the electronic device based on the response sound data.

3. The method of claim 1, further comprising,

in response to a determination that the keyword data file is compatible with the keyword detection unit, extracting the keyword model from the keyword data file and storing the keyword model.

4. The method of claim 1, wherein the second version of the keyword model is compatible with the keyword detection unit and further comprising storing the second version of the keyword model.

5. The method of claim 1, further comprising, in response to a determination that the keyword data file is not compatible with the keyword detection unit:

sending information descriptive of the electronic device, the keyword detection unit, or both, via the communication network to a server, wherein the server selects the second version of the keyword model based on the information, wherein the second version of the keyword model is compatible with the keyword detection unit;

receiving the second version of the keyword model from the server; and

storing the second version of the keyword model at the electronic device.

6. The method of claim 1, further comprising, in response to a determination that the keyword data file is not compatible with the keyword detection unit, converting the keyword model to the second version, wherein the second version of the keyword model is compatible with the keyword detection unit and storing the second version of the keyword model.

7. The method of claim 1, wherein assigning the particular target keyword to the function comprises combining the keyword model with at least one other keyword model associated with a second particular target keyword from the plurality of target keywords to form a combined keyword model, and wherein the second particular target keyword is associated with at least one other function of the electronic device.

8. The method of claim 7, further comprising:

receiving the input sound at the electronic device;

detecting, at the electronic device, whether the particular target keyword, the second particular target keyword, or a combination thereof, is included in the input sound based on the combined keyword model;

performing the function in response to detecting that the particular target keyword is included in the input sound; and

performing the at least one other function in response to detecting that the second particular target keyword is included in the input sound.

9. The method of claim 1, wherein the keyword data file includes an application file, and further comprising extracting the keyword model from the application file, wherein the function to which the particular target keyword is assigned includes activating the application file.

10. The method of claim 1, further comprising receiving a voice tone model associated with the particular target keyword, wherein the voice tone model is adapted to modify a voice tone of speech output by the electronic device.

11. The method of claim 10, wherein the voice tone model is adapted to modify the voice tone of speech that is converted from text, and further comprising:

receiving the input sound at the electronic device;

detecting the particular target keyword in the input sound based on the keyword model; and

in response to detecting that the particular target keyword is included in the input sound, modifying the voice tone of the speech converted from the text based on the voice tone model associated with the particular target keyword.

12. The method of claim 1, further comprising:

receiving the input sound at the electronic device;

detecting the particular target keyword in the input sound based on the keyword model; and

in response to detecting that the particular target keyword is included in the input sound, activating the function, wherein activating the function comprises executing instructions causing the electronic device to perform the function.

13. The method of claim 1, wherein the function comprises activating a voice assistant, activating a camera, activating a music player, activating a voice recorder, taking a photograph, unlocking the electronic device, or a combination thereof.

14. The method of claim 1, further comprising receiving a selection of the function via a user interface of the electronic device prior to assigning the particular target keyword to the function.

15. An electronic device comprising:

a user interface unit to receive input selecting a particular target keyword from a list of a plurality of target keywords;

a download management unit configured to receive, via a communication network in response to the input, a keyword data file, the keyword data file including a keyword model for the particular target keyword and including compatibility information;

a verification unit configured to determine whether the keyword data file is compatible with a keyword detection unit based on the compatibility information, wherein the download management unit is configured to receive a second version of the keyword model based on a determination that the keyword data file is not compatible with the keyword detection unit; and

a keyword setting unit configured to assign the particular target keyword to a function of the electronic device in response to the input such that the function is activated in response to detecting, based on the keyword model, that an input sound received at the electronic device includes the particular target keyword.

16. The electronic device of claim 15, wherein the download management unit is further configured to receive response sound data, the response sound data used to generate a response sound indicative of detection of the particular target keyword is included in the input sound.

17. The electronic device of claim 16, further comprising:

a sound sensor configured to receive the input sound, wherein the keyword detection unit is configured to detect whether the particular target keyword is included in the input sound based on the keyword model; and

a function management unit configured to generate the response sound based on the response sound data associated with the particular target keyword in response to detecting that the particular target keyword is included in the input sound.

18. The electronic device of claim 15, wherein the download management unit is further configured to receive a voice tone model adapted to modify a voice tone of speech converted from text.

19. The electronic device of claim 18, further comprising:

a sound sensor configured to receive the input sound;

a keyword detection unit configured to detect that the particular target keyword is included in the input sound based on the keyword model; and

a text-to-speech (TTS) conversion unit configured to modify the voice tone of the speech converted from the text based on the voice tone model in response to detection of the particular target keyword in the input sound.

20. The electronic device of claim 15, wherein the keyword setting unit is further configured to combine the keyword model with at least one other keyword model associated with a second particular target keyword to form a combined keyword model, wherein the second particular target keyword is associated with at least one other function of the electronic device.

21. The electronic device of claim 20, further comprising:

a sound sensor configured to receive the input sound;

a keyword detection unit configured to detect whether keywords are included in the input sound based on the combined keyword model; and

a function management unit configured to;

activate the function in response to detection of the particular target keyword in the input sound based on the combined keyword model; and

activate the at least one other function in response to detection of the second particular target keyword in the input sound based on the combined keyword model.

22. The electronic device of claim 15, wherein the download management unit is configured to receive, via the communication network, an application file, wherein the application file includes the keyword data file, and further comprising an extracting unit configured to extract the keyword data file from the application file, wherein the keyword setting unit is configured to assign the particular target keyword to the application file for activating the application in response to detecting the particular target keyword in the input sound based on the keyword model.

23. The electronic device of claim 22, further comprising:

a sound sensor configured to receive the input sound;

a keyword detection unit configured to detect whether the particular target keyword is included in the input sound based on the keyword model; and

a function management unit configured to activate the application in response to detecting the particular target keyword is included in the input sound.

24. The electronic device of claim 15, further comprising:

a sound sensor configured to receive the input sound;

a keyword detection unit configured to detect whether the particular target keyword is included in the input sound based on the keyword model; and

a function management unit configured to perform the function in response to detecting that the particular target keyword is included in the input sound.

25. An electronic device comprising:

means for receiving input selecting a particular target keyword from a list of a plurality of target keywords;

means for receiving, via a communication network in response to the particular target keyword, a keyword data file, the keyword data file including a keyword model for the particular target keyword and including compatibility information;

means for determining, based on the compatibility information, whether the keyword data file is compatible with means for detecting keywords in an input sound, wherein the means for receiving the keyword data file is configured to receive a second version of the keyword model based on a determination that the keyword data file is not compatible with the means for detecting keywords in the input sound; and

means for assigning the particular target keyword to a function in response to the input such that the function is activated in response to detecting, based on the keyword model, that the input sound received at the electronic device includes the particular target keyword.

26. The electronic device of claim 25, wherein the means for receiving the keyword model is configured to receive response sound data, wherein the means for detecting keywords is configured to determine whether the particular target keyword is included in the input sound and further comprising:

means for receiving the input sound;

and

means for generating a response sound based on the response sound data associated with the particular target keyword in response to detecting that the particular target keyword is included in the input sound.

27. The electronic device of claim 25, further comprising:

means for extracting the keyword model from the keyword data file in response to a determination that the keyword data file is compatible with the means for detecting keywords in the input sound.

28. The electronic device of claim 27, wherein the second version of the keyword model is compatible with the means for detecting keywords and further comprising storing the second version of the keyword model.

29. A non-transitory computer-readable storage medium storing instructions, the instructions causing a processor to:

receive, at an electronic device, a list of a plurality of target keywords, the list received via a communication network;

receive user input that selects a particular target keyword from the list of the plurality of target keywords; and

in response to the user input;

receive a keyword data file via the communication network, the keyword data file including a keyword model for the particular target keyword and including compatibility information;

determine whether the keyword data file is compatible with the electronic device based on the compatibility information;

receive a second version of the keyword model based on a determination that the keyword data file is not compatible with a keyword detection unit; and

assigning, at the electronic device, the particular target keyword to a function of the electronic device such that the function is activated in response to detecting, based on the keyword model, that an input sound received at the electronic device includes the particular target keyword.

30. The non-transitory computer-readable storage medium of claim 29, wherein the keyword model is non-user specific and includes information to model a subset of phonemes of a language of the particular target keyword.