System and method of supporting adaptive misrecognition in conversational speech

• US 9,626,959 B2
• Filed: 12/30/2013
• Issued: 04/18/2017
• Est. Priority Date: 08/10/2005
• Status: Active Grant

• Current Assignee: Dialect, LLC
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Di Cristo, Philippe, Weider, Chris, Kennewick, Robert A.
• Primary Examiner(s): He, Jialong
• Application Number:
  US14/143,903
• Publication Number:
  US 20140108013A1
• Time in Patent Office: 1,205 Days
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A system and method are provided for receiving speech and/or non-speech communications of natural language questions and/or commands and executing the questions and/or commands. The invention provides a conversational human-machine interface that includes a conversational speech analyzer, a general cognitive model, an environmental model, and a personalized cognitive model to determine context, domain knowledge, and invoke prior information to interpret a spoken utterance or a received non-spoken message. The system and method creates, stores, and uses extensive personal profile information for each user, thereby improving the reliability of determining the context of the speech or non-speech communication and presenting the expected results for a particular question or command.

Claims

1. A method of processing natural language command, the method being implemented by a computer system that comprises one or more physical processors executing one or more computer program instructions which, when executed, perform the method, the method comprising:

receiving, by the computer system, a natural language command from a user;

generating, by the computer system, a first interpretation of the natural language command based on one or more recognized words of the natural language command;

performing, by the computer system, a first action specified by the natural language command based on the first interpretation;

accessing, by the computer system, a personalized cognitive model to proactively select a second interpretation of the natural language command responsive to an indication from the user that the first interpretation is not correct; and

proactively performing, by the computer system, a second action specified by the natural language command based on the second interpretation.

2. The method of claim 1, further comprising:

receiving, by the computer system, a second natural language command from the user; and

generating, by the computer system, one or more interpretations of the second natural language command,

wherein proactively performing, by the computer system, a second action comprises performing the second action based on the one or more interpretations of the second natural language command.

3. The method of claim 1, further comprising:

updating, by the computer system, the personalized cognitive model associated with the user based on the indication that the first interpretation is not correct,

wherein proactively performing, by the computer system, the second action comprises performing the second action based on the updated personalized cognitive model.

4. The method of claim 1, further comprising:

receiving, by the computer system, a user input associated with the user after the receipt of the natural language command,

wherein the indication from the user that the first interpretation is not correct is based on a determination that the receipt of the user input is proximate in time to the receipt of the natural language command.

5. The method of claim 4, wherein the determination that the receipt of the user input is proximate in time to the receipt of the natural language command is based on a determination that the user input is received before a predetermined time.

6. The method of claim 4, wherein the user input includes a subsequent natural language command or a non-voice input.

7. The method of claim 1, further comprising:

receiving, by the computer system, a user input associated with the user after the receipt of the natural language command; and

determining, by the computer system, whether the receipt of the user input is proximate in time to the receipt of the natural language command; and

determining, by the computer system, whether the first interpretation is correct, wherein determining whether the first interpretation is correct comprises;

determining that the first interpretation is correct in response to a determination that the receipt of the user input is not proximate in time to the receipt of the natural language command; and

determining that the first interpretation is not correct in response to a determination that the receipt of the user input is proximate in time to the receipt of the natural language command.

8. The method of claim 1, wherein the second action includes an action that is predicted to be requested or taken by the user after the receipt of the natural language command.

9. A system for processing natural language utterances, comprising:

one or more physical processors programmed to execute one or more computer program instructions which, when executed, cause the one or more physical processors to;

receive a natural language command associated with a user;

generate a first interpretation of the natural language command based on one or more recognized words of the natural language command;

perform a first action specified by the natural language command based on the first interpretation;

access, by the computer system, a personalized cognitive model to proactively select a second interpretation of the natural language command responsive to an indication from the user that the first interpretation is not correct; and

proactively perform a second action specified by the natural language command based on the second interpretation.

10. The system of claim 9, wherein the one or more physical processors are further caused to:

receive a second natural language command associated with the user; and

generate one or more interpretations of the second natural language command,

perform the second action based on the one or more interpretations of the second natural language command.

11. The system of claim 9, wherein the one or more physical processors are further caused to:

update the personalized cognitive model associated with the user based on the indication that the first interpretation is not correct,

wherein proactively perform the second action based on the second interpretation comprises performing the second action based on the updated personalized cognitive model.

12. The system of claim 9, wherein the one or more physical processors are further caused to:

receive a user input associated with the user after the receipt of the natural language command,

wherein the indication from the user that the first interpretation is not correct is based on a determination that the receipt of the user input is proximate in time to the receipt of the natural language command.

13. The system of claim 12, wherein the determination that the receipt of the user input is proximate in time to the receipt of the natural language command is based on a determination that the user input is received before a predetermined time.

14. The system of claim 12, wherein the user input includes a subsequent natural language command or a non-voice input.

15. The system of claim 9, wherein the one or more physical processors are further caused to:

receive a user input associated with the user after the receipt of the natural language command; and

determine whether the receipt of the user input is proximate in time to the receipt of the natural language command; and

determine whether the first interpretation is correct, wherein determine whether the interpretation is correct comprises;

determine that the first interpretation is correct in response to a determination that the receipt of the user input is not proximate in time to the receipt of the natural language command; and

determine that the first interpretation is not correct in response to a determination that the receipt of the user input is proximate in time to the receipt of the natural language command.

16. The system of claim 9, wherein the second action includes an action that is predicted to be requested or taken by the user after the receipt of the natural language command.