Multi computational agent performance of tasks

Patent metadata

Field	Value
Publication number	US-11663535-B2
Application number	US-201715815368-A
Country	US
Kind code	B2
Filing date	Nov 16, 2017
Priority date	Oct 3, 2016
Publication date	May 30, 2023
Grant date	May 30, 2023

Abstract

An example method includes receiving, by one or more processors, a representation of an utterance spoken at a computing device; identifying, by a first computational agent from a plurality of computational agents and based on the utterance, a multi-element task to be performed, wherein the plurality of computational agents includes one or more first party computational agents and a plurality of third-party computational agents; and performing, by the first computational agent, a first sub-set of elements of the multi-element task, wherein performing the first sub-set of elements comprises selecting a second computational agent from the plurality of computational agents to perform a second sub-set of elements of the multi-element task.

First claim

The invention claimed is:

1. A method implemented by one or more processors of a first party computational agent, the method comprising:
receiving a representation of an utterance spoken at a computing device;
identifying, based on the utterance, a multi-element task to be performed, the multi-element task including at least a first sub-set of elements and a second sub-set of elements;
identifying, the first party computational agent to perform the multi-element task;
in response to determining that a capability level of the first party computational agent indicates the first party computational agent can only perform the first sub-set of elements of the multi-element task:
determining capability levels of respective third party computational agents, of a plurality of third party computational agents, to perform the multi-element task;

determining that a particular capability level of a particular third party computational agent, of the plurality of third party computational agents, indicates the particular third party computational agent can perform the second sub-set of elements of the multi-element task;
in response to determining that the particular capability level of the particular third party computational agent indicates the particular third party computational agent can perform the second sub-set of elements of the multi-element task:
selecting based on an agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to an occurrence of the utterance associated with at least the second sub-set of elements, the particular third party computational agent, from among the plurality of third party computational agents, to perform the second sub-set of elements of the multi-element task;

causing the first party computational agent to perform the first sub-set of elements of the multi-element task;
causing the particular third party computational agent to perform the second sub-set of elements of the multi-element task;
subsequent to causing the particular third party computational agent to perform the second sub-set of elements of the multi-element task:
receiving, from the particular third party computational agent, an indication of actions performed by the particular third party computational agent; and
updating, based on the indication of the actions performed by the particular third party computational agent, the agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to future occurrences of the utterance associated with at least the second sub-set of elements; and

subsequent to updating the agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to the future occurrences of the utterance associated with at least the second sub-set of elements:
receiving an additional representation of the utterance spoken at the computing device or an additional computing device;
identifying, based on the utterance, an additional multi-element task to be performed, the additional multi-element task including at least the second sub-set of elements; and
utilizing the updated the updated agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to the additional utterance associated with at least the second sub-set of elements in determining whether to select the particular third party computational agent or an additional particular third party computational agent, from among the plurality of third party computational agents, to perform the second sub-set of elements.

2. The method of claim 1 , wherein performing the first sub-set of elements further comprises:
determining that additional information is needed to perform the second sub-set of elements;
causing the first party computational agent to gather the additional information; and
causing the first party computational agent to output the gathered additional information to the particular third party computational agent.

3. The method of claim 2 , wherein causing the first party computational agent to output the gathered additional information to the particular third party computational agent comprises:
responsive to determining, based on an authorization data store, that the particular third party computational agent is authorized to receive the gathered additional information, causing the first party computational agent to output the gathered additional information to the particular third party computational agent.

4. The method of claim 1 , wherein the one or more processors of the first party computational agent are included in the computing device.

5. The method of claim 1 , wherein the one or more processors of the first party computational agent are included in a distributed computing system.

6. The method of claim 1 , wherein causing the particular third party computational agent performs the second sub-set of elements of the multi-element task comprises:
causing the particular third party computational agent to perform the second sub-set of elements of the multi-element task in parallel with causing the first party computational agent to perform the first sub-set of elements of the multi-element task.

7. The method of claim 1 , further comprising:
determining, based on at least historical usage of the particular third party computational agent, the agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to the occurrence of the utterance associated with at least the second sub-set of elements.

8. The method of claim 1 , further comprising:
obtaining, from an agent index, the agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to an occurrence of the utterance associated with at least the second sub-set of elements,
wherein updating the agent quality value that is assigned to the particular third party computational agent and that influences whether the particular third party computational agent is selected responsive to future occurrences of the utterance associated with at least the second sub-set of elements based on the indication of the actions performed by the particular third party computational agent comprises causing the agent quality value to be updated in the agent index.

9. A computing device comprising:
at least one processor of a first party computational agent; and
at least one memory comprising instructions that when executed, cause the at least one processor of the first party computational agent to:
receive a representation of an utterance spoken at a computing device;
identify, based on the utterance, a multi-element task to be performed, the multi-element task including at least a first sub-set of elements and a second sub-set of elements;
identify the first party computational agent to perform the multi-element task;
in response to determining that a capability level of the first party computational agent indicates the first party computational agent can only perform the first sub-set of elements of the multi-element task:
determine capability levels of respective third party computational agents, of a plurality of third party computational agents, to perform the multi-elem

Assignees

• Google Llc

Inventors

• Stets Robert
• Nygaard Valerie
• Caprita Bogdan
• Abrams Bradley M
• Douglas Jason Brant

Classifications

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G10L15/22

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

• G06Q10/063112 Primary

  Skill-based matching of a person or a group to a task · CPC title