Service for developing dialog-driven applications

• US 10,331,791 B2
• Filed: 11/23/2016
• Issued: 06/25/2019
• Est. Priority Date: 11/23/2016
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Anbazhagan, Vikram Sathyanarayana, Pokkunuri, Rama Krishna Sandeep, Sivasubramanian, Swaminathan, Stefani, Stefano, Zhukov, Vladimir
• Primary Examiner(s): Vo, Huyen X
• Application Number:
  US15/360,814
• Publication Number:
  US 20180143967A1
• Time in Patent Office: 944 Days
• Field of Search: 704 1- 10, 704230-257, 704270-2701, 704275
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: G06F 8/30 Creation or generation of source code
• CPC Class Codes: G06N 20/00 Machine learning
• 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 natural language understanding model is trained using respective natural language example inputs corresponding to a plurality of applications. A determination is made as to whether a value of a first parameter of a first application is to be obtained using a natural language interaction. Using the natural language understanding model, at least a portion of the first application is generated.

Claims

1. A system, comprising:

one or more processors; and

memory storing program instructions that, if executed, cause the one or more processors to perform a method comprising;

training a natural language understanding model using at least respective natural language input examples corresponding to a plurality of applications;

determining whether a value of a first parameter of a first application is to be obtained using a natural language interaction; and

based, at least in part, on determining that the value of the first parameter of the first application is to be obtained using the natural language interaction, generating, using the natural language understanding model, at least a portion of the first application associated with obtaining the value of the first parameter.

2. The system as recited in claim 1, wherein the method comprises:

storing an indication that, in response to a determination that an end user has approved a use of profile-based personalization with respect to the first application, a value of the first parameter is to be selected based at least in part on a profile record of the end user.

3. The system as recited in claim 1, wherein the natural language interaction comprises one or more of:

(a) a speech-based interaction or (b) a text-based interaction.

4. The system as recited in claim 1, wherein the method comprises:

identifying respective resources of one or more network-accessible services of a provider network, wherein the respective resources are to be used to implement the first application; and

storing an indication of the respective resources.

5. The system as recited in claim 1, wherein the method comprises:

including, within a data set used for training the natural language understanding model, one or more records of end-user interactions with at least one application.

6. A method, comprising:

training a natural language understanding model using respective natural language input examples corresponding to a plurality of applications;

determining whether a value of a first parameter of a first application is to be obtained using a natural language interaction; and

based, at least in part, on determining that the value of the first parameter of the first application is to be obtained using the natural language interaction, generating, using the natural language understanding model, at least a portion of the first application associated with obtaining the value of the first parameter.

7. The method as recited in claim 6, further comprising:

determining that an end user has approved a use of profile-based personalization with respect to the first application; and

selecting a value of the first parameter based at least in part on a profile record of the end user.

8. The method as recited in claim 6, further comprising:

receiving, via a graphical interface of a network-accessible service of a provider network, at least one natural language input example of the respective natural language input examples.

9. The method as recited in claim 6, further comprising:

receiving an indication of respective resources of one or more network-accessible services of a provider network, wherein the respective resources are to be used to implement the first application; and

storing the indication of the respective resources.

10. The method as recited in claim 6, wherein the first application comprises a plurality of tasks, and wherein the first parameter is associated with the first task, the method further comprising:

storing an indication that, based at least in part on a determination that at least a portion of a first natural language interaction has been completed, a second natural language interaction to determine a value of one or more parameters of the second task is to be initiated.

11. The method as recited in claim 6, further comprising:

storing an indication that a range of permissible values of a second parameter of the first application depends at least in part on a value of the first parameter; and

generating a natural language word string to indicate, to an end user of the first application, the range of permissible values.

12. The method as recited in claim 6, wherein the natural language interaction comprises one or more of:

(a) a speech-based interaction or (b) a text-based interaction.

13. The method as recited in claim 6, further comprising:

storing one or more log records indicative of respective interactions with respective end-users of the first application.

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

in response to a query received via a programmatic interface, providing an indication of the one or more log records.

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

adding, to a data set used for training the natural language understanding model, the one or more log records.

16. A non-transitory computer-accessible storage medium storing program instructions that when executed on one or more processors cause the one or more processors to:

train a natural language understanding model using respective natural language input examples corresponding to a plurality of applications;

determine whether a value of a first parameter of a first application is to be obtained using a natural language interaction; and

based, at least in part, on the determination that the value of the first parameter of the first application is to be obtained using the natural language interaction, generate using the natural language understanding model, at least a portion of the first application associated with obtaining the value of the first parameter.

17. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the instructions when executed on the one or more processors cause the one or more processors to:

store an indication that, subsequent to a determination that an end user has approved a use of profile-based personalization with respect to the first application, a value of the first parameter is to be selected based at least in part on a profile record of the end user.

18. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the instructions when executed on the one or more processors cause the one or more processors to:

determine, using a programmatic interface of a network-accessible service of a provider network, a first natural language input example of the respective natural language input examples.

19. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the instructions when executed on the one or more processors cause the one or more processors to:

store an indication that respective resources of one or more network-accessible services of a provider network are to be used to implement the first application.

20. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the natural language interaction comprises one or more of:

(a) a speech-based interaction or (b) a text-based interaction.