Systems And Methods For Managing Interactions From Multiple Speech-Enabled Applications

• US 20070143115A1
• Filed: 12/16/2005
• Published: 06/21/2007
• Est. Priority Date: 02/04/2002
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Miller, David, Yip, Clement, Banay, Dan, Falcon, Stephen
• Granted Patent: US 8,374,879 B2
• Time in Patent Office: Days
• US Class Current: 704/270.100
• CPC Class Codes: G01C 21/26 specially adapted for navigation in a road network
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

Systems and methods are described for speech systems that utilize an interaction manager to manage interactions—also known as dialogues—from one or more applications. The interactions are managed properly even if multiple applications use different grammars. The interaction manager maintains an interaction list. An application wishing to utilize the speech system submits one or more interactions to the interaction manager. Interactions are normally processed in the order in which they are received. An exception to this rule is an interaction that is configured by an application to be processed immediately, which causes the interaction manager to place the interaction at the front of the interaction list of interactions. If an application has designated an interaction to interrupt a currently processing interaction, then the newly submitted application will interrupt any interaction currently being processed and, therefore, it will be processed immediately.

Claims

1. A speech system, comprising:

a speech server;

an interaction manager configured to prioritize interactions submitted to the speech server by ordering the interactions in an interaction list, the interaction list having a front and an end, wherein an interaction at the front of the interaction list is processed before other interactions in the interaction list are processed;

wherein the interaction manager places an interaction at the end of the interaction list unless the interaction manager detects an indication to place the interaction at the front of the interaction list ahead of other interactions in the interaction list; and

wherein the speech server identifies a speech-enabled application associated with each interaction at the front of the interaction list, and if the associated speech-enabled application is not loaded, then the speech server loads the speech-enabled application.

2. The speech system as recited in claim 1, wherein the interaction manager places the interaction at the front of the list if the interaction manager detects an interruption flag in the interaction.

3. The speech system as recited in claim 2, wherein the interaction manager interrupts an interaction currently processing when the interaction is placed at the front of the list, and the interrupted interaction resumes processing after the interrupting interaction is processed.

4. The speech system as recited in claim 2, wherein the interaction manager interrupts an interaction currently processing when the interaction is placed at the front of the list, and the interrupted interaction does not resume processing after the interrupting interaction is processed.

5. The speech system as recited in claim 2, wherein the interaction manager does not interrupt an interaction currently processing when the interaction is placed at the front of the list, but processes the interrupting interaction after the currently processing interaction has completed processing if the currently processing interaction will conclude processing in less than a predetermined period of time.

6. The speech system as recited in claim 2, wherein the interaction manager only interrupts an interaction currently processing if the currently processing interaction will not conclude processing for more than a predetermined period of time.

7. The speech system as recited in claim 1, further comprising a speech engine that communicates with the speech server through a speech application programming interface (API).

8. The speech system as recited in claim 1, wherein the interaction manager provides a grace period after processing a first interaction before beginning to process a second interaction.

9. A method for processing speech interactions associated with multiple speech-enabled applications, comprising:

receiving an interaction;

determining if the interaction is an interrupting interaction;

placing the interaction in an interaction list having a front and a back and containing from one to several interactions to be processed in order from the front to the back, wherein the interaction is placed at the back of the interaction list unless the interaction is an interrupting interaction, in which case the interaction is placed at the front of the interaction list;

processing the interaction when it is at the front of the list, including identifying one of the speech-enabled applications associated with the interaction at the front of the list and if the speech-enabled application is not loaded, then loading the speech-enabled application.

10. The method as recited in claim 9, wherein the interaction list is a list of pointers to memory locations that contain the interactions submitted to the interaction list.

11. The method as recited in claim 9, wherein the determining if the interaction is an interrupting interaction further comprises checking for the presence of an interruption flag in the interaction that, if present, indicates that the interaction is an interrupting interaction.

12. The method as recited in claim 9, wherein the interaction is a first interaction and the method further comprises waiting a predetermined grace period after the first interaction is processed before beginning processing of a second interaction.

13. The method as recited in claim 12, further comprising:

receiving a third interaction during the grace period; and

processing the third interaction prior to processing the second interaction.

14. One or more computer-readable media containing computer executable instructions that, when executed on a computer, perform the following steps:

maintaining an interaction list that includes multiple speech interactions;

identifying a front of the interaction list and a back of the interaction list, the interactions in the interaction list to be processed in order from front to back;

processing a first interaction when the first interaction is located at the front of the interaction list;

receiving a second interaction from a second speech-enabled application;

determining if the second interaction is an interrupting interaction;

if the second interaction is not an interrupting interaction, placing the second interaction at the back of the interaction list;

if the second interaction is an interrupting interaction, placing the second interaction at the front of the interaction list and processing the second interaction; and

identifying a speech-enabled application associated with each interaction at the front of the interaction list, and if the associated speech-enabled application is not loaded, then loading the speech-enabled application.

15. The one or more computer-readable media as recited in claim 14, wherein the first interaction is associated with a first grammar and the second interaction is associated with a second grammar.