Centralized method and system for determining voice commands

• US 10,748,530 B2
• Filed: 10/13/2016
• Issued: 08/18/2020
• Est. Priority Date: 11/16/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Mowatt, David, Chambers, Robert L., Andrew, Felix G. T. I.
• Primary Examiner(s): Ortiz-Sanchez, Michael
• Application Number:
  US15/292,871
• Publication Number:
  US 20170032786A1
• Time in Patent Office: 1,405 Days
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A computing system includes at least one processor and memory storing instructions that, when executed, cause the computing system to receive an indication of a set of grammar commands for an application associated with the computing system, each grammar command corresponding to an executable command that is executable by the application to perform a corresponding application action, receive an indication of a first speech input, from a user, captured by a speech capture component, detect an ambiguity based on a correlation between the first speech input and the set of grammar commands, provide, to the user, an output that represents the detected ambiguity, receive, in response to the output, an indication of a second speech input that is captured by the speech capture component and resolves the ambiguity, and, based on the second speech input execute the executable command that corresponds to one of the grammar commands.

Claims

1. A method performed by a computing system, the method comprising:

receiving, by a grammar generator, a first input corresponding to a first application that is associated with the computing system,

the first input identifying a set of terms associated with a first executable command, that is executable to perform a corresponding application action in the first application;

receiving, by the grammar generator, a second input corresponding to a second application that is associated with the computing system,

the second input identifying a set of terms associated with a second executable command, that is executable to perform a corresponding application action in the second application;

based on the first and second inputs, generating, by the grammar generator, a grammar that is stored in a data store and includes grammar commands that correspond to the first and second executable commands;

in response to an indication of a first user speech input, accessing, by a module that is separate from the first and second applications, the grammar stored in the data store;

determining, by the module, a first correlation between terms in the first user speech input and a first one of the grammar commands that corresponds to the first executable command;

determining, by the module a second correlation between the terms in the first user speech input and a second one of the grammar commands that corresponds to the second executable command;

detecting an ambiguity based on the first and second correlations;

providing an output that represents the detected ambiguity;

receiving, in response to the output, an indication of a second user speech input;

based on the second user speech input, selecting, by the module, an executable command from the first executable command and the second executable command; and

generating an instruction that

launches one of the first application or the second application that corresponds to the selected executable command; and

performs the corresponding application action in the launched application.

2. The method of claim 1, further comprising:

storing each grammar command in association with a corresponding one of the first or second applications.

3. The method of claim 1, wherein the grammar commands include a set of grammar commands that correspond to alternative forms of the first executable command.

4. The method of claim 3, wherein the alternative forms of the first executable grammar command have a same definition of the first executable command.

5. The method of claim 1, and further comprising:

rendering an indication to a user that identifies at least the first executable command; and

based on the second user speech input,

selecting the first executable command; and

launching the first application to perform the corresponding application action in the first application.

6. The method of claim 5, wherein the indication identifies the first application.

7. The method of claim 1, wherein the terms are determined by a speech recognizer processing the first user speech input.

8. A computing system comprising:

at least one processor; and

memory storing instructions executable by the at least one processor, wherein the instructions, when executed, cause the computing system to;

receive an indication of a set of grammar commands for an application associated with the computing system, each grammar command corresponding to an executable command that is executable by the application to perform a corresponding application action; and

receive an indication of a first speech input, from a user, captured by a speech capture component;

determine a first correlation between terms in the first speech input and a first one of the grammar commands that corresponds to a first executable command;

determine a second correlation between terms in the first speech input and a second one of the grammar commands that corresponds to a second executable command;

detect an ambiguity based on the first and second correlations;

provide, to the user, an output that represents the detected ambiguity;

receive, in response to the output, an indication of a second speech input that is captured by the speech capture component and resolves the ambiguity; and

based on the second speech input, execute the executable command that corresponds to one of the grammar commands.

9. The computing system of claim 8, wherein the corresponding one of the grammar commands is selected based at least in part on one or more terms identified from the second speech input.

10. The computing system of claim 8, wherein the output includes an application identifier that is rendered to the user and identifies the application.

11. The computing system of claim 10, wherein two or more of the plurality of grammar commands correspond to alternative forms of a particular one of the executable commands that have the same definition of the particular executable command.

12. The computing system of claim 8, wherein the instructions, when executed, cause the computing system to launch the application to execute the selected executable command.

13. A computing system comprising:

at least one processor; and

memory storing instructions executable by the at least one processor, wherein the instructions, when executed, configure the computing system to;

receive, a first input corresponding to a first application that is associated with the computing system,

the first input identifying a set of terms associated with a first set of executable commands, each being executable by the first application to perform a corresponding application action;

receive a second input corresponding to a second application that is associated with the computing system,

the second input identifying a set of terms associated with a second set of executable commands, each being executable by the second application to perform a corresponding application action;

generate a grammar that is stored in a data store and includes a plurality of grammar commands based on the first and second inputs from the first and second applications, wherein each grammar command;

corresponds to one of the executable commands, and

is stored in association with a respective one of the first or second applications;

based on a first user speech input from a user, access the grammar using a module that is separate from the first and second applications;

determine a first correlation between terms in the first user speech input and a first one of the grammar commands;

determine a second correlation between terms in the first user speech input and a second one of the grammar commands;

detect an ambiguity based on the first and second correlations;

provide an output that represents the detected ambiguity;

receive, in response to the output, an indication of a second user speech input;

based on the second user speech input, select one of the first or second grammar commands; and

generate an instruction that launches the application associated with the selected grammar command and executes the executable command corresponding to the selected grammar command.

14. The computing system method of claim 13, wherein the selected grammar command is selected based on a correlation of the second user speech input to the grammar command.

15. The computing system method of claim 13, wherein two or more of the plurality of grammar commands correspond to alternative forms of a particular one of the executable commands and have the same definition of the particular executable command.

16. The computing system method of claim 13, wherein the output includes an indication of the first grammar command.

17. The computing system method of claim 16, wherein the output comprises a prompt that identifies the first application.

18. The computing system method of claim 13, wherein the instructions configure the computing system to:

define an operational instruction based at least in part on the selected executable command and a portion correlated to the first user speech input.

19. The computing system method of claim 13, wherein the instructions configure the computing system to:

identify one or more executable commands that correspond to the first user speech input; and

determine that the first user speech input is ambiguous based on at least one of;

a determination that the one or more executable commands that can be executed to perform more than one action;

or

a determination that the one or more executable commands can be executed by more than one of the applications.