Speech recognition system loading different recognition engines for different applications

• US 7,203,645 B2
• Filed: 04/27/2001
• Issued: 04/10/2007
• Est. Priority Date: 04/27/2001
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Pokhariyal, Shuvranshu, Aundhe, Shirish, Hernandez, Thomas
• Primary Examiner(s): Opsasnick; Michael N.
• Application Number:
  US09/844,288
• Publication Number:
  US 20030014251A1
• Time in Patent Office: 2,174 Days
• Field of Search: 704/251, 704/275
• US Class Current: 704/251
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A speech recognition framework receives information about a recognized phrase from a speech engine. The framework identifies the application that is a focus of the recognized phrase. The framework then selects a handler function based on the recognized phrase and the application. The framework calls the handler function, which responds to the phrase.

Claims

1. A method comprising:

receiving information about which of a plurality of applications is being currently used;

automatically selecting, based on said information, one among multiple sets of handling information, each set of handling information being associated with a different of said applications and loading a first grammar for a first application based on said automatically selecting where said first grammar is associated with, but separate from, the first application and detecting a change in said information and unloading said first grammar, and loading a second, different grammar for a second application on the speech engine, where said grammar is separate from the second application.

2. A method as in claim 1, wherein said receiving comprises receiving information about a recognized phrase that includes a wildcard portion which is filled in with a parameter from a selected application.

3. The method of claim 1, wherein said recurring information comprises:

identifying an application that is a focus of the recognized phrase, and said automatically selecting being based on the identified application.

4. The method of claim 3 wherein said automatically selecting comprises:

selecting a set of handling information based on the identified application; and

recognizing selecting a handler function from the selected set of handling information based on the recognized phrase.

5. The method of claim 4 further comprising, prior to receiving the recognized phrase:

locating a plurality of sets of handling information.

6. The method of claim 5 further comprising loading the speech engine, the first application, and the second application on a user'"'"'s computer, and wherein each of the sets of handling information is located when the execution of the associated application is initiated.

7. The method of claim 6 further comprising:

directing an operating system to provide notification whenever the execution of an application is initiated;

wherein each set of handling information is located when the notification is provided.

8. The method of claim 5 further comprising:

detecting a change of the focus from a first application to a second application;

producing the second grammar based on the handling information associated with the second application; and

loading the second grammar onto the speech engine.

9. The method of claim 8 further comprising:

generating an uncompiled grammar based on the handling information; and

compiling the grammar into a binary format.

10. The method of claim 8 further comprising:

directing an operating system to provide notification in response to the focus changing;

wherein the step of determining when the focus changes includes receiving notification from an operating system.

11. The method of claim 8 further comprising:

storing the produced grammar; and

loading the stored grammar onto the speech engine when the focus is changed from a third application to the second application.

12. An article comprising a machine-readable medium which stores machine-executable instructions, the instructions causing a machine to:

receive information about which of a plurality of applications is being currently used; and

select, based on said information, one set of handling information, each set of handling information being associated with a different of said applications and load a first grammar for a first application based on said automatically select, where first grammar is associated with, but separate from the first application and detect a change in said information and unload said first grammar, and load a second, different grammar for a second application on the speech engine, where said grammar is separate from the second application.

13. An article as in claim 12, wherein said receiving comprises receiving information about a recognized phrase that includes a wildcard portion which is filled in with a parameter from a selected application.

14. The article of claim 12, wherein the instruction further cause the machine to:

identify an application that is a focus of the recognized phrase, selecting the handler function being further based on the identified application.

15. The article of claim 14 wherein selecting a handler function comprises:

selecting a set of handling information based on the identified application; and

selecting a handler function from the selected set of handling information based on the recognized phrase.

16. The article of claim 15 wherein the instructions further cause the machine to:

generate an uncompiled grammar based on the handling information; and

compile the grammar into a binary format.

17. The article of claim 15 wherein the instructions further cause the machine, prior to receiving the recognized phrase, to:

locate sets of handling information, each of the sets of handling information being associated with a different application.

18. The article of claim 17 wherein each of the sets of handling information is located when the execution of the associated application is initiated.

19. The article of claim 18 wherein the instructions further cause the machine to:

direct an operating system to provide notification whenever the execution of an application is initiated;

wherein each set of handling information is located when the notification is provided.

20. The article of claim 17 wherein the instructions further cause the machine to:

detect a change of the focus from a first application to a second application;

produce a second grammar based on the handling information associated with the second application; and

load the second grammar onto the speech engine.

21. The article of claim 20 wherein the instructions further cause the machine to:

direct an operating system to provide notification in response to the focus changing;

wherein the step of determining when the focus is changed includes receiving notification from an operating system that the focus has been changed.

22. An apparatus comprising:

a memory which stores computer readable instructions;

a processor which executes the computer readable instructions, the instructions causing the processor to;

receive information about a recognized phrase from a speech engine;

identity an application that is a focus of the recognized phrase; and

select a handler function based on the recognized phrase and the application that is the focus of the phrase and loading a first grammar for a first application based on said select, where said first grammar is separate from the first application, and detecting a change in said information and unloading said first grammar, and loading a second, different grammar for a second application on the speech engine, where said second grammar is separate from the second application.

23. An apparatus as in claim 22, wherein the processor receives the information about the recognized phrase, that includes a wildcard portion which is filled with a parameter from a selected application.

24. The apparatus of claim 22 wherein selecting a handler function comprises:

selecting a set of handling information based on the identified application; and

selecting a handler function from the selected set of handling information based on the recognized phrase.

25. The apparatus of claim 24 wherein the instructions further cause the processor to:

generate an uncompiled grammar based on the handling information; and

compile the grammar into a binary format.

26. The apparatus of claim 24 wherein the instructions further cause the processor, prior to receiving the recognized phrase, to:

locate sets of handling information, each of the sets of handling information being associated with a different application.

27. The apparatus of claim 26 wherein each of the sets of handling information is located when the execution of the associated application is initiated.

28. The apparatus of claim 27 wherein the instructions further cause the processor to:

direct an operating system to provide notification whenever the execution of an application is initiated;

wherein each set of handling information is located when the notification is provided.

29. The apparatus of claim 26 wherein the instructions further cause the processor to:

detect a change of the focus from a first application to a second application;

produce a second grammar based on the handling information associated with the second application; and

load the second grammar onto the speech engine.

30. The apparatus of claim 29 wherein the instructions further cause the processor to:

direct an operating system to provide notification in response to the focus changing;

wherein the step of determining when the focus is changed includes receiving notification from an operating system that the focus has been changed.