Voice user interface authoring tool

• US 20070156406A1
• Filed: 04/11/2006
• Published: 07/05/2007
• Est. Priority Date: 12/30/2005
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Odell, Julian J., Ollason, David G., Barton, William F., Spina, Michelle S.
• Granted Patent: US 8,315,874 B2
• Time in Patent Office: Days
• US Class Current: 704/257
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A voice user interface authoring tool is configured to use categorized example caller responses, from which callflow paths, automatic speech recognition, and natural language processing control files can be generated automatically within a single, integrated authoring user interface. A voice user interface (VUI) design component allows an author to create an application incorporating various types of action nodes, including Prompt/Response Processing (PRP) nodes. At runtime, the system uses the information from each PRP node to prompt a user to say something, and to process the user'"'"'s response in order to extract its meaning. An Automatic Speech Recognition/Natural Language Processing (ASR/NLP) Control Design component allows the author to associate sample inputs with each possible meaning, and automatically generates the necessary ASR and NLP runtime control files. The VUI design component allows the author to associate the appropriate ASR and NLP control files with each PRP node, and to associate an action node with each possible meaning, as indicated by the NLP control file.

Claims

1. A method implemented by a computing device, comprising:

enabling an author to define one or more prompt nodes and one or more response classes, and to assign sample inputs to each of the response classes;

generating one or more automatic speech recognition control files and natural language processing control files, based on the sample inputs, for the response classes; and

generating flow paths for the response classes from the prompt nodes.

2. The method of claim 1, further comprising enabling an author to configure one or more response containers and to assign one or more of the response classes to the response containers.

3. The method of claim 2, further comprising enabling an author to associate one or more of the response containers with a corresponding one of the prompt nodes.

4. The method of claim 2, wherein a response container associated with one of the prompt nodes contains a response class that contains a keyword container comprising one or more keyword values, and one of the flow paths is generated for each of the keywords.

5. The method of claim 2, further comprising assigning the automatic speech recognition and natural language processing control files to the response containers.

6. The method of claim 1, further comprising assigning the response classes to connect to action nodes.

7. The method of claim 6, wherein the action nodes comprise one or more nodes associated with an announcement action.

8. The method of claim 6, wherein the action nodes comprise one or more nodes associated with a transaction action.

9. The method of claim 6, wherein the action nodes comprise one or more additional prompt nodes.

10. The method of claim 1, wherein the automatic speech recognition control files and natural language processing control files are combined into unified control files.

11. The method of claim 1, wherein the natural language processing control files are created with a classifier for associating sample inputs with the response classes.

12. A method implemented by a computing device, comprising:

receiving one or more sample inputs;

enabling each of the sample inputs to be associated with a response class;

automatically generating automatic speech recognition control files and natural language processing control files from the sample inputs for the response classes;

enabling one or more response containers, prompt/response processing nodes, and action nodes to be specified;

enabling the response classes to be assigned to the one or more response containers, and the one or more response containers to be assigned to the one or more prompt/response processing nodes; and

automatically assigning one or more of the response classes as flow paths from the prompt/response processing nodes to the action nodes.

13. The method of claim 12, wherein the automatic speech recognition control files and the natural language processing control files are imported into a voice user interface by value.

14. The method of claim 12, wherein the automatic speech recognition control files and the natural language processing control files are imported into a voice user interface by reference.

15. The method of claim 12, further comprising using learned sample inputs to generate revised automatic speech recognition control files component and revised automatic speech recognition component control files.

16. The method of claim 12, further comprising using format rules to generate one or more of the natural language processing control files.

17. The method of claim 12, further comprising:

measuring characteristics of the sample inputs; and

using the characteristics to create a classifier configured for providing an indication of which of the response classes corresponds to subsequent speech inputs.

18. The method of claim 17, further comprising evaluating a level of confidence for the indication of which of the response classes corresponds to subsequent speech inputs.

19. The method of claim 18, further comprising providing one or more options for corrective action if the level of confidence for the indication of which of the response classes corresponds to a subsequent speech input is below a selected threshold.

20. A medium comprising instructions readable by a computing device, the instructions enabling the computing device to execute a voice user interface authoring tool, comprising:

an automatic speech recognition component;

a natural language processing component; and

a voice user interface design component, enabling one or more flow paths to be specified between one or more prompt nodes and one or more action nodes, and configured to import control files from the natural language processing component and the automatic speech recognition component based on sample inputs associated with the response classes.