Correcting speech recognition without first presenting alternatives

• US 6,347,296 B1
• Filed: 06/23/1999
• Issued: 02/12/2002
• Est. Priority Date: 06/23/1999
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Friedland, Steven J.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): Nolan, Daniel A
• Application Number:
  US09/338,699
• Time in Patent Office: 965 Days
• Field of Search: 704/270, 704/272, 704/278, 704/231, 704/251, 704/252
• US Class Current: 704/231
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A method and a system within a computer speech recognition system to automatically correct portions of text. The invention receives text derived from a first user input for inclusion in a body of text. Concurrently upon receipt of the first user input, and based upon the first user input, the system identifies a list of alternate text selections potentially intended by the user. Next, the system stores each of the alternate text selections in a memory location associated with the text. In response to a second user input, and without displaying the list of alternate text selections to the user, the system automatically retrieves a first one of the alternate text selections from the memory location and inserts the first one of the alternate text selections in place of the text in the body of text.

Claims

1. In a computer speech recognition system, a method of automatically correcting portions of text comprising:

receiving text derived from a first user input for inclusion in a body of text;

concurrently upon receipt of said first user input, and based upon said first user input, identifying alternate text selections potentially intended by said user;

storing each of said alternate text selections in a memory location associated with said text;

in response to a second user input identifying at least a portion of said text, and without displaying said alternate text selections to said user, automatically retrieving a first one of said alternate text selections from said memory location and inserting said first one of said alternate text selections in place of said at least a portion of said text in said body of text.

2. The method according to claim 1 further comprising the step of:

in response to a third user input, automatically replacing said first alternate text selection in said body of text with a second one of said alternate text selections.

3. The method according to claim 2 further comprising the step of:

in response to a fourth user input, replacing said second alternate text selection in said body of text with at least one of said text and said alternate text selections, which has previously been included in said body of text.

4. The method according to claim 1 wherein each of said user inputs is a spoken utterance.

5. The method according to claim 4 wherein said text and each of said alternate text selections are derived from said spoken utterance by a computer speech recognition engine.

6. The method according to claim 1 wherein said text is derived from a user keyboard entry.

7. The method according to claim 1 whereby said second user input includes:

selecting said text to be replaced; and

articulating a spoken command for requesting replacement of said text with one of said alternate text selections.

8. In a computer speech recognition system, a system for automatically correcting portions of text comprising:

means for receiving text derived from a first user input for inclusion in a body of text;

means for identifying alternate text selections potentially intended by said user concurrently upon receipt of said first user input, and based upon said first user input;

means for storing each of said alternate text selections in a memory location associated with said text;

means for automatically retrieving a first one of said alternate text selections from said memory location, and inserting said first one of said alternate text selections in place of at least a portion of said text in said body of text, in response to a second user input identifying said at least a portion of said text, and without displaying said alternate text selections to said user.

9. The system according to claim 8 further comprising:

means for automatically replacing said first alternate text selection in said body of text with a second one of said alternate text selections in response to a third user input.

10. The system according to claim 9 further comprising:

means for replacing said second alternate text selection in said body of text with at least one of said text and said alternate text selections, which has previously been included in said body of text, in response to a fourth user input.

11. The system according to claim 8 wherein each of said user inputs is a spoken utterance.

12. The system according to claim 11 whereby said means for receiving said text and identifying each of said alternate text selections is a computer speech recognition engine.

13. The system according to claim 8 whereby said text is derived from a user keyboard entry.

14. The system according to claim 8 whereby said means of automatically retrieving a first one of said alternate text selections and inserting said first one of said alternate text selections in place of said text in said body of text includes:

means for said user to select said text in said body of text; and

means of requesting, by articulating a spoken command, replacement of said text in said body of text with one of said alternate text sections.

15. 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:

receiving text derived from a first user input for inclusion in a body of text;

concurrently upon receipt of said first user input, and based upon said first user input, identifying alternate text selections potentially intended by said user;

storing said alternate text selections in a memory location associated with said text;

in response to a second user input identifying at least a portion of said text, and without displaying said alternate text selections to said user, retrieving a first one of said alternate text selections from said memory location and replacing said at least a portion of said text with said first one of said alternate text selections.

16. The machine readable storage according to claim 15 for causing the machine to perform the further step of:

in response to a third user input, replacing said first alternate text selection in said body of text with a second one of said alternate text selections.

17. The machine readable storage according to claim 16 for causing the machine to perform the further step of:

in response to a fourth user input, replacing said second alternate text selection in said body of text with at least one of said text and said alternate text selections which has previously been included in said body of text.

18. The machine readable storage according to claim 15 causing the machine to perform the further step of:

deriving said text and each of said alternate text selections from a spoken utterance by a computer speech recognition engine.

19. The machine readable storage according to claim 15 causing the machine to perform the further step of:

deriving said text from a user keyboard entry.