PERFORMING A SAFETY ANALYSIS FOR USER-DEFINED VOICE COMMANDS TO ENSURE THAT THE VOICE COMMANDS DO NOT CAUSE SPEECH RECOGNITION AMBIGUITIES

• US 20080046250A1
• Filed: 07/26/2006
• Published: 02/21/2008
• Est. Priority Date: 07/26/2006
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: VILA, ROBERTO, AGAPI, CIPRIAN, BLASS, OSCAR J., MONTEIRO, BRENNAN D.
• Granted Patent: US 8,234,120 B2
• Time in Patent Office: Days
• US Class Current: 704/275
• CPC Class Codes: G10L 15/075 supervised, i.e. under machine guidance

Abstract

The present invention discloses a solution for assuring user-defined voice commands are unambiguous. The solution can include a step of identifying a user attempt to enter a user-defined voice command into a voice-enabled system. A safety analysis can be performed on the user-defined voice command to determine a likelihood that the user-defined voice command will be confused with preexisting voice commands recognized by the voice-enabled system. When a high likelihood of confusion is determined by the safety analysis, a notification can be presented that the user-defined voice command is subject to confusion. A user can then define a different voice command or can choose to continue to use the potentially confusing command, possibly subject to a system imposed confusion mitigating condition or action.

Claims

1. A method of assuring user-defined voice commands are unambiguous comprising:

identifying a user attempt to enter a user-defined voice command into a voice-enabled system;

performing a safety analysis on the user-defined voice command to determine a likelihood that the user-defined voice command will be confused with preexisting voice commands recognized by the voice-enabled system; and

when a high likelihood of confusion is determined by the safety analysis, presenting a notification that the user-defined voice command is subject to confusion.

2. The method of claim 1, further comprising:

when a low likelihood of confusion is determined by the safety analysis, immediately accepting the new user-defined voice command as a new command recognized by the voice-enabled system.

3. The method of claim 1, wherein the new user-defined voice command is associated with a user-defined abstraction.

4. The method of claim 3, further comprising:

accepting the new user-defined voice command as a command for the user-defined abstraction after the safety analysis is performed;

after the accepting step, recording a plurality of actions associated with the user-defined abstraction; and

storing the plurality of actions and relating these actions to the user-defined voice command so that the user-defined voice command is able to be used to invoke an execution of the plurality of actions in the future.

5. The method of claim 1, further comprising:

providing an option to enter a different user-defined voice command, when a high likelihood of confusion is determined by the safety analysis.

6. The method of claim 1, further comprising:

when a high likelihood of confusion is determined by the safety analysis, presenting a warning message to a user that the user-defined command is potentially able to be confused with a different voice command; and

permitting the user to retain the user-defined command.

7. The method of claim 6, further comprising:

requiring the user to preface the user-defined command with a keyword before being invoked, where user-defined commands that do not have a high likelihood of confusion as determined by the safety analysis are able to be invoked without being prefaced with a keyword.

8. The method of claim 1, further comprising:

when a high likelihood of confusion is determined by the safety analysis and when a user elects to continue using the new user-defined voice command, performing at least one programmatic action to minimize recognition ambiguities between the preexisting voice command and the new user-defined voice command; and

accepting the new user-defined voice command as a new command recognized by the voice-enabled system.

9. The method of claim 1, further comprising:

parsing the new user-defined command into a plurality of component parts, wherein the safety analysis is performed against each component part.

10. The method of claim 1, wherein the steps of claim 1 are performed by at least one of a service agent and a computing device manipulated by the service agent, the steps being performed in response to a service request.

13. The method of claim 11, wherein the steps of claim 1 are performed by at least one of a service agent and a computing device manipulated by the service agent, the steps being performed in response to a service request.

11. A method of adding user-defined abstractions to a voice-enabled system comprising:

detecting an action to record a new abstraction;

receiving a user-defined voice command for invoking the new abstraction;

comparing the user-defined voice command against a set of preexisting voice commands to determine whether a potential recognition ambiguity exists between the user-defined voice command and the preexisting voice commands;

when a potential recognition ambiguity exists, performing at least one programmatic action designed to minimize the likelihood of an occurrence of the potential recognition ambiguity;

recording a plurality of actions associated with the new abstraction; and

accepting the new abstraction and the associated user-defined voice command, wherein the user-defined voice command is able to be used to invoke an execution of the plurality of actions in the future.

12. The method of claim 11, wherein said at least one programmatic action comprises an action selected from a set of actions consisting of substituting a different user-defined voice command for the user-defined voice command, accepting the user-defined voice command yet limiting a recognition of the command to situations where the command is prefaced with a keyword, and placing the user-defined voice command in a grammar different from a grammar containing potentially ambiguous preexisting voice commands where the grammars are contextually dependent grammars that are not simultaneously activated by the voice-enabled system.

14. A voice-enabled system comprising:

a grammar of user-defined voice commands, each voice command being associated with at least one programmatic action;

a speech recognition engine configured to accept speech, to compare the speech with the voice commands in the grammar, and to automatically execute programmatic actions associated with the speech commands responsive to a match between the speech and the voice commands; and

a safety analysis engine configured to determine a likelihood that new user-defined voice commands will be confused with the voice commands in the grammar, wherein the safety analysis engine is utilized for each new user-defined voice command before that new user-defined voice command is permitted to be added to the grammar.

15. The system of claim 14, further comprising:

an abstraction engine configured to record and execute user-defined abstractions., each user-defined abstraction being associated with one of the user-defined voice commands in the grammar, and wherein said at least one programmatic action comprises user-defined actions associated with the user-defined abstraction.

16. The system of claim 14, further comprising:

at least one grammar of system-defined voice commands, wherein the voice commands processed by the speech recognition engine and the safety analysis engine include the system-defined voice commands.

17. The system of claim 14, further comprising:

a confusion mitigation engine configured to automatically execute at least one programmatic action designed to minimize a likelihood of confusion between new user-defined voice commands and preexisting voice commands.

18. The system of claim 17, wherein the confusion mitigation engine presents a notification to a user that a new user-defined voice command is subject to confusion whenever a high likelihood of confusion is determined by the safety analysis.

19. The system of claim 17, wherein the confusion mitigation engine requires a new user-defined voice command having a high likelihood of confusion as determined by the safety analysis to be preceded by a keyword in order to be invoked, wherein user-defined voice commands having a low likelihood of confusion as determined by the safety analysis are able to be directly invoked without the keyword.

20. The system of claim 17, wherein the confusion mitigation engine automatically restricts a user set able to utilize a new user-defined voice command to prevent potential conflicts with other user-defined voice commands that are not available to the restricted user set but are available to other users of the voice-enabled system.