Apparatus and method for voice command recognition based on a combination of dialog models

• US 8,954,326 B2
• Filed: 09/26/2011
• Issued: 02/10/2015
• Est. Priority Date: 01/04/2011
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd. (Samsung Group)
• Original Assignee: Samsung Electronics Co. Ltd. (Samsung Group)
• Inventors: Kwak, Byung-Kwan, Park, Chi-Youn, Kim, Jeong-Su, Cho, Jeong-Mi
• Primary Examiner(s): GUERRA-ERAZO, EDGAR X
• Application Number:
  US13/245,032
• Publication Number:
  US 20120173244A1
• Time in Patent Office: 1,233 Days
• Field of Search: 704/275, 704/235, 704/246, 704/270, 704/270.1
• US Class Current: 704/246
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Provided are a voice command recognition apparatus and method capable of figuring out the intention of a voice command input through a voice dialog interface, by combining a rule based dialog model and a statistical dialog model rule. The voice command recognition apparatus includes a command intention determining unit configured to correct an error in recognizing a voice command of a user, and an application processing unit configured to check whether the final command intention determined in the command intention determining unit comprises the input factors for execution of an application.

Claims

1. An apparatus for voice command recognition based on a combination of dialog models, the apparatus comprising:

a command intention determining unit configured to recognize a command intention of a voice command of a user through a command intention probability distribution corresponding to the voice command of a user; and

an application processing unit configured to recognize the command intention to select an application corresponding to the command intention, and to generate an execution signal of the application.

2. The apparatus of claim 1, wherein the command intention determining unit is configured to correct an error in recognizing a voice command of a user through a command intention probability distribution corresponding to the voice command of a user, and configured to determine a final command intention of the voice command, and wherein the command intention determining unit comprises:

a command intention probability distribution update unit configured to update the command intention probability distribution corresponding to the voice command of the user;

an error determining unit configured to determine the error in recognizing the voice command or configured to determine an error in figuring out the command intention through an updated command intention probability distribution; and

a re-input requesting response generating unit configured to request re-input of the voice command in response to the error determining unit determining that an error occurs.

3. The apparatus of claim 2, wherein the command intention determining unit comprises a final command intention selecting unit configured to select the final command intention in the updated command intention probability in response to the error determining unit determining that an error does not occur.

4. The apparatus of claim 3, wherein the command intention determining unit comprises a selection requesting response generating unit configured to generate a response requesting selection of a candidate for the final command intention of the updated command intention probability distribution.

5. The apparatus of claim 3, wherein the command intention determining unit comprises a confirmation requesting response generating unit configured to generate a request for requesting confirmation of the final command intention of the updated command intention probability distribution.

6. The apparatus of claim 2, wherein an application processing unit configured to check whether the final command intention comprises all input factors for execution of an application, to acquire information related to one or more missing input factors from the user, and to generate an execution signal of the application corresponding to the final command intention, and wherein the application processing unit comprises:

an input factor checking unit configured to check whether the final command intention comprises all of the input factors for execution of the application; and

a user intention adding unit configured to request one or more missing input factors in response to one or more of the input factors for execution of the application missing from the final command intention.

7. The apparatus of claim 6, wherein the application processing unit further comprises:

a reliability checking unit configured to check whether the input factor is an input factor that is available to execute the application in response to the final command intention comprising all the input factors for executing the application; and

a user intention checking unit configured to request re-input corresponding to an unavailable input factor in response to the unavailable input factor being an input factor for executing the application.

18. The apparatus of claim 1, wherein the application processing unit comprises an application rule storage unit configured to store an application rule for checking with the final command intention.

8. A portable communication device comprising:

a voice interface unit configured to receive a voice command signal corresponding to a voice command of a user, and to output a voice response;

a voice command recognition unit configured to recognize a command intention of the voice command through a command intention probability distribution corresponding to the voice command signal that is input through the voice interface to recognize the command intention, to select an application corresponding to the command intention, and to generate an execution signal of the appication; and

an operating unit configured to operate the application according to the application execution signal that is generated in the voice command recognition unit.

9. The portable communication device of claim 8, further comprising a communication unit configured to perform a wired/wireless communication with a web server, wherein the portable communication device downloads related information from the web server through the communication unit according to the application execution signal that is generated in the voice command recognition unit.

10. A method of voice command recognition based on a combination of dialog models, the method comprising:

correcting an error in recognizing a voice command of a user through a command intention probability distribution of a command intention corresponding to the voice command;

determining a final command intention of the voice command;

checking whether the determined final command intention comprises all of the input factors for execution of an application;

acquiring information related to one or more missing input factors from the user; and

generating an execution signal of the application corresponding to the final command intention.

11. The method of claim 10, wherein the determining of the final command intention comprises:

updating the command intention probability distribution corresponding to the voice command of the user;

determining the error in recognizing the voice command or an error in figuring out the command intention through a updated command intention probability distribution; and

requesting re-input of the voice command in response to a result of the determination being that an error occurs.

12. The method of claim 10, wherein the generating of the execution signal of the application comprises:

checking whether the final command intention comprises all of the input factors for execution of the application; and

requesting one or more missing input factors in response to one or more of the input factors for execution of the application missing from the final command intention.

13. The method of claim 12, wherein the generating of the execution signal of the application comprises:

checking whether each of the input factors is an input factor available to execute the application in response to the final command intention comprising all the input factors for executing the application; and

requesting re-input corresponding to an unavailable input factor in response to the unavailable input factor being an input factor for executing the application.

14. A command intention determiner comprising:

an update unit configured to generate and to update a command intention probability distribution based on dialog spoken by a user;

an error determining unit configured to compare an input voice command to the command intention probability distribution to determine whether an error recognition occurs in the input voice command; and

a request unit configured to request a user to repeat an input voice command, in response to the comparison performed by the error determining unit determining that an error occurs in the input voice command.

15. The command intention determining unit of claim 14, wherein the command intention probability distribution is calculated through a statistical dialog modeling.

16. The command intention determining unit of claim 14, further comprising a command intention selecting unit configured to select a command intention that has a probability that exceeds a threshold value in the command intention probability distribution, in response to the comparison performed by the error determining unit determining that an error does not occur in the input voice command.

17. The command intention determining unit of claim 16, further comprising a confirmation request unit,

wherein the command intention selecting unit is further configured to determine whether a plurality of command intentions have a probability that exceeds the threshold value, and the confirmation request unit requests the user for a confirmation of the selected command intention, in response to the plurality of command intentions exceeding the threshold value.