Method of dynamically displaying speech recognition system information

• US 7,099,829 B2
• Filed: 11/06/2001
• Issued: 08/29/2006
• Est. Priority Date: 11/06/2001
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Gomez, Felipe
• Primary Examiner(s): Hudspeth, David
• Assistant Examiner(s): Sked, Matthew J.
• Application Number:
  US09/992,816
• Publication Number:
  US 20030088413A1
• Time in Patent Office: 1,757 Days
• Field of Search: 704/231, 704/275, 704/235
• US Class Current: 704/275
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

The present invention provides a method of dynamically displaying speech recognition system information. The method can include providing a single floating window for displaying frames of speech recognition system state information to a user. The frames can be varied according to trigger events detected in the speech recognition system. Each frame can differ from others of the frames according to the speech recognition system state information.

Claims

1. A method of dynamically displaying speech recognition system information comprising:

providing a single floating window for displaying frames of speech recognition system state information to a user;

performing an empirical analysis of a plurality of user actions and inputs to determine a plurality of different trigger events that cause operating states of the speech recognition system to change in response thereto; and

dynamically updating said single floating window for individually and sequentially displaying said frames at different times, each individual display changing in response to said different trigger events detected in said speech recognition system, wherein each said individually displayed frame differs from others of said frames according to said speech recognition system state information, wherein the frames variably and individually displayed in the single floating window include a frame containing a list of valid speech recognition commands for a current speech recognition system state and a frame containing a list of alternative text selections for a previously spoken word for which a speech-recognition operation has been performed, wherein each variably and individually displayed frame is uniquely associated with a specific trigger event, and wherein at least one of said trigger events is an automatic user-independent event.

2. The method of claim 1, further comprising:

detecting a trigger event in said speech recognition system; and

responsive to said trigger event, dynamically updating said single floating window according to said trigger event.

3. The method of claim 2, wherein each of said trigger events specifies one of a plurality of context dependent frames.

4. The method of claim 3, said updating step comprising:

changing said context dependent frame in said single floating window to a context dependent frame corresponding to said detected trigger event, wherein said context dependent frame includes selected items of speech recognition system state information.

5. The method of claim 4, further comprising:

determining at least one of said items of speech recognition system state information to be included in said context dependent frame corresponding to said detected trigger event.

6. The method of claim 5, wherein at least one trigger event comprises one of a user selection of text in a primary view of said speech recognition system, a user command to initiate a function in said primary view, and a location of a pointer in said primary view.

7. The method of claim 5, wherein said selected items of speech recognition state information include a list of available speech commands, a list of alternate text selections, and a list of commands previously issued by a user.

8. A system for providing speech recognition system information, the system comprising:

an empirical analyzer for performing an empirical analysis of a plurality of system user actions and inputs to determine a plurality of different trigger events that cause operating states of the speech recognition system to change in response thereto; and

a single graphical user interface configured to display all context dependent frames of selected items of speech recognition system state information in said speech recognition system, wherein said single graphical user interface is further configured to dynamically and individually present at different times selected ones of said plurality of context dependent frames in response to said different trigger events detected in said speech recognition system, wherein at least one trigger event comprises an automatic user-independent event, wherein the context dependent frames dynamically presented in the single graphical user interface include a frame containing a list of valid speech recognition commands for a current speech recognition system state and a frame containing a list of alternative text selections for a previously spoken word for which a speech-recognition operation has been performed; and

wherein each frame presented is uniquely associated with a specific trigger event and individually presented separate from other frames.

9. The graphical user interface of claim 8, wherein said selected items of speech recognition system state information include a list of available speech commands, a list of alternate text selections, and a list of commands previously issued by a user.

10. The graphical user interface of claim 8, wherein at least one trigger event comprises one of a change in an operational state of a speech recognition system, a user selection of text, a user command, and a location of a pointer.

11. A machine-readable storage, having stored thereon a computer program having a plurality of code sections executable by a machine for causing the machine to perform the steps of:

providing a single floating window for displaying frames of speech recognition system state information to a user;

performing an empirical analysis of a plurality of user actions and inputs to determine a plurality of different trigger events that cause operating states of the speech recognition system to change in response thereto; and

dynamically updating said single floating window for individually and sequentially displaying said frames at different times, each individual display changing in response to said different trigger events detected in said speech recognition system, wherein each said individually displayed frame differs from others of said frames according to said speech recognition system state information, wherein the frames variably and individually displayed in the single floating window include a frame containing a list of valid speech recognition commands for a current speech recognition system state and a frame containing a list of alternative text selections for a previously spoken word for which a speech-recognition operation has been performed, wherein each variably and individually displayed frame is uniquely associated with a specific trigger event, and wherein at least one of said trigger events is an automatic user-independent event.

12. The machine-readable storage of claim 11, said updating step comprising;

detecting a trigger event in said speech recognition system; and

responsive to said trigger event, dynamically updating said single floating window according to said trigger event.

13. The machine-readable storage of claim 12, wherein each of said trigger events specifies one of a plurality of context dependent frames.

14. The machine-readable storage of claim 13, said updating step comprising:

changing said context dependent frame in said single floating window to a context dependent frame corresponding to said detected trigger event, wherein said context dependent frame includes selected items of speech recognition system state information.

15. The machine-readable storage of claim 14, further comprising:

determining at least one of said items of speech recognition system state information to be included in said context dependent frame corresponding to said detected trigger event.

16. The machine-readable storage of claim 15, wherein at least one trigger event comprises one of a change in an operational state of said speech recognition system, a user selection of text in a primary view of said speech recognition system, a user command to initiate a function in said primary view, and a location of a pointer in said primary view.

17. The machine-readable storage of claim 15, wherein said selected items of speech recognition system state information include available speech commands, a list of alternate text selections, and a list of commands previously issued by a user.