Alternate response generation

Patent metadata

Field	Value
Publication number	US-12165671-B2
Application number	US-202318507715-A
Country	US
Kind code	B2
Filing date	Nov 13, 2023
Priority date	Dec 10, 2018
Publication date	Dec 10, 2024
Grant date	Dec 10, 2024

Abstract

Techniques for performing conversation recovery of a system/user exchange are described. In response to determining that an action responsive to a user input cannot be performed, a system may determine a topic to recommend to a user. The topic may be unrelated to the original substance of the user input. The system may have access to various data representing a context in which a user provides an input to the system. The system may use these inputs and various data at runtime to make a determination regarding whether a user should be recommended a topic, as well as what that topic should be. The system may cause a question be output to the user, with the question asking the user about the topic, for example whether the user would like a song played, whether the user would like to hear information about a particular individual (e.g., artist), whether the user would like to know about a particular skill (e.g., a skill having a significantly high popularity among users of the system), or whether the user would like to know about some other topic. If the user responds affirmatively to the recommended topic, the system may pass the user experience off to an appropriate component of the system (e.g., one that is configured to perform an action related to the topic). If the user responds negatively, does not respond at all, or the system is unsure whether the user's response was affirmative or negative, the system may cease interaction with the user, thereby enabling the user to interact with the system as the user desires.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
receiving, from a first device, first data representing a first natural language input;
determining the first data corresponds to a user identifier;
determining an error condition in response to the first natural language input;
determining second data representing at least one previous natural language input corresponding to the user identifier, wherein the at least one previous natural language input relates to natural language processing data associated with the user identifier and/or an action previously performed based on the natural language processing data associated with the user identifier;
using the second data to determine, from a plurality of techniques for recovering from the error condition, a technique for recovering from the error condition;
using the technique to determine output data corresponding to a response to the first natural language input; and
causing the output data to be sent to the first device.

2. The computer-implemented method of claim 1 , wherein:
the error condition corresponds to selecting from a plurality of potential interpretations of the first natural language input;
the technique corresponds to selection of a first potential interpretation of the plurality of potential interpretations, the selection corresponding to the at least one previous natural language input; and
using the technique corresponds to using the first potential interpretation.

3. The computer-implemented method of claim 1 , wherein the second data corresponds to at least one processing component enabled with respect to the user identifier.

4. The computer-implemented method of claim 1 , wherein the error condition corresponds to a potential error.

5. The computer-implemented method of claim 1 , wherein the error condition corresponds to a potential recommendation to be output.

6. The computer-implemented method of claim 1 , wherein the second data corresponds to a previous output topic recommendation.

7. The computer-implemented method of claim 1 , wherein the technique corresponds to outputting a prompt corresponding to a topic.

8. The computer-implemented method of claim 1 , wherein the technique corresponds to outputting a prompt corresponding to an action to be performed in response to a natural language input.

9. The computer-implemented method of claim 1 , further comprising:
processing the first data to determine a first potential hypothesis corresponding to a first confidence value and a second potential hypothesis corresponding to a second confidence value, wherein the first confidence value indicates a higher confidence that the first potential hypothesis corresponds to the first natural language input than the second potential hypothesis;
determining that the technique corresponds to selecting a hypothesis that does not correspond to a higher confidence value; and
performing processing with regard to the second potential hypothesis to determine the output data.

10. The computer-implemented method of claim 1 , further comprising:
determining the at least one previous natural language input occurred within a certain amount of time from the first natural language input.

11. A system comprising:
at least one processor; and
at least one memory comprising instructions that, when executed by the at least one processor, cause the system to:
receive, from a first device, first data representing a first natural language input;
determine the first data corresponds to a user identifier;
determine an error condition in response to the first natural language input;
determine second data representing at least one previous natural language input corresponding to the user identifier, wherein the at least one previous natural language input relates to natural language processing data associated with the user identifier and/or an action previously performed based on the natural language processing data associated with the user identifier;
use the second data to determine, from a plurality of techniques for recovering from the error condition, a technique for recovering from the error condition;
use the technique to determine output data corresponding to a response to the first natural language input; and
cause the output data to be sent to the first device.

12. The system of claim 11 , wherein:
the error condition corresponds to selecting from a plurality of potential interpretations of the first natural language input;
the technique corresponds to selection of a first potential interpretation of the plurality of potential interpretations, the selection corresponding to the at least one previous natural language input; and
use of the technique corresponds to use of the first potential interpretation.

13. The system of claim 11 , wherein the second data corresponds to at least one processing component enabled with respect to the user identifier.

14. The system of claim 11 , wherein the error condition corresponds to a potential error.

15. The system of claim 11 , wherein the error condition corresponds to a potential recommendation to be output.

16. The system of claim 11 , wherein the second data corresponds to a previous output topic recommendation.

17. The system of claim 11 , wherein the technique corresponds to outputting a prompt corresponding to a topic.

18. The system of claim 11 , wherein the technique corresponds to outputting a prompt corresponding to an action to be performed in response to a natural language input.

19. The system of claim 11 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
process the first data to determine a first potential hypothesis corresponding to a first confidence value and a second potential hypothesis corresponding to a second confidence value, wherein the first confidence value indicates a higher confidence that the first potential hypothesis corresponds to the first natural language input than the second potential hypothesis;
determine that the technique corresponds to selecting a hypothesis that does not correspond to a higher confidence value; and
perform processing with regard to the second potential hypothesis to determine the output data.

20. The system of claim 11 , wherein the at least one memory further comprises instructions that, when executed by the at least one processor, further cause the system to:
determine the at least one previous natural language input occurred within a certain amount of time from the first natural language input.

Assignees

• Amazon Tech Inc
• Amazon Tech

Inventors

• Newell Gregory
• Kahan Eliav
• Yasa Ravi Chandra Reddy
• Suarez David
• Toledano Joel

Classifications

• G10L2015/225

  Feedback of the input speech · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L15/26

  Speech to text systems (G10L15/08 takes precedence) · CPC title

• G10L15/22

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G10L15/1822

  Parsing for meaning understanding · CPC title