Conversational agent response determined using a sentiment or user profile data

Patent metadata

Field	Value
Publication number	US-9997158-B2
Application number	US-201715464935-A
Country	US
Kind code	B2
Filing date	Mar 21, 2017
Priority date	Jul 31, 2014
Publication date	Jun 12, 2018
Grant date	Jun 12, 2018

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for handing off a user conversation between computer-implemented agents. One of the methods includes receiving, by a computer-implemented agent specific to a user device, a digital representation of speech encoding an utterance, determining, by the computer-implemented agent, that the utterance specifies a requirement to establish a communication with another computer-implemented agent, and establishing, by the computer-implemented agent, a communication between the other computer-implemented agent and the user device.

First claim

What is claimed is:

1. A system comprising a user device and one or more storage devices on which are stored instructions that are operable, when executed by the user device, to cause the user device to perform operations comprising:
receiving, by a computer-implemented agent specific to the user device, a digital representation of speech encoding an utterance;
in response to processing the digital representation of the speech to determine the words included in the utterance, determining, using the utterance, to request input identifying a sentiment of a user who spoke the utterance;
in response to determining to request the input identifying a sentiment of the user who spoke the utterance, presenting a request for user input identifying a sentiment;
in response to presenting the request, receiving user input identifying a sentiment of a speaker of the utterance;
determining, by the computer-implemented agent, that the utterance specifies a request for another computer-implemented agent that is a different computer-implemented agent than the computer-implemented agent by:
determining, by the computer-implemented agent using the utterance, that the utterance is not specific to a particular second computer-implemented agent;
in response to determining that the utterance is not specific to a particular second computer-implemented agent, determining, by the computer-implemented agent using the utterance, two or more second computer-implemented agents each of which can respond to the utterance;
in response to determining the two or more second computer-implemented agents each of which can respond to the utterance, automatically selecting, without receiving user input, one of the second computer-implemented agents each of which can respond to the utterance; and
in response to automatically selecting, without receiving user input, one of the second computer-implemented agents each of which can respond to the utterance, determining, by the computer-implemented agent, that the utterance specifies a request for the selected one of the second computer-implemented agents;

providing, by the computer-implemented agent to the other computer-implemented agent, a representation of the utterance and data identifying the sentiment and user profile data for a user who spoke the utterance to enable the other computer-implemented agent to automatically complete a transaction using the user profile data without requesting, from the user, data that is included in the user profile data;
receiving, by the user device, a response to the utterance for the other computer-implemented agent that was determined using the sentiment; and
in response to receiving identification of the response, presenting, by the user device, the response to the utterance received from the other computer-implemented agent that was determined using the sentiment.

2. The system of claim 1 , wherein presenting the request for user input identifying the sentiment comprises requesting user input identifying a yes or a no response to a question.

3. The system of claim 1 , wherein presenting the request for user input that identifying the sentiment comprises requesting user input identifying an option from a set of options.

4. The system of claim 1 , wherein receiving the response to the utterance from the other computer-implemented agent comprises:
receiving, by the user device from the other computer-implemented agent, another representation of a response to the utterance; and
generating, using the other representation, data to cause a presentation of the response to the utterance.

5. The system of claim 1 wherein the other computer-implemented agent is not specific to the user device.

6. The system of claim 1 wherein the computer-implemented agent is for a first application on the user device and the other computer-implemented agent is for a second, different application on the user device.

7. The system of claim 1 comprising:
data, stored on the user device, that represents the other computer-implemented agent and includes text-to-speech parameters for responses provided by the other computer-implemented agent, the operations comprising:
determining, by the other computer-implemented agent implemented on the user device using the sentiment and the data that represents the other computer-implemented agent, the response, wherein:
presenting, by the user device, the response to the utterance received from the other computer-implemented agent that was determined using the sentiment comprises presenting, by the user device, the response using the text-to-speech parameters.

8. The system of claim 1 comprising:
receiving, by the user device, a description of an entity that created the other computer-implemented agent; and
providing, by the user device, a representation of the description of the entity that created the other computer-implemented agent for presentation.

9. The system of claim 1 , wherein selecting the one of the second computer-implemented agents comprises selecting, using the user profile data for the user who spoke the utterance, the one of the second computer-implemented agents.

10. A computer-implemented method comprising:
receiving, by a computer-implemented agent specific to a user device, a digital representation of speech encoding an utterance;
in response to processing the digital representation of the speech to determine the words included in the utterance, determining, using the utterance, to request input identifying a sentiment of a user who spoke the utterance;
in response to determining to request the input identifying a sentiment of the user who spoke the utterance, presenting a request for user input identifying a sentiment;
in response to presenting the request, receiving user input identifying a sentiment of a speaker of the utterance;
determining, by the computer-implemented agent, that the utterance specifies a request for another computer-implemented agent that is a different computer-implemented agent than the computer-implemented agent by:
determining, by the computer-implemented agent using the utterance, that the utterance is not specific to a particular second computer-implemented agent;
in response to determining that the utterance is not specific to a particular second computer-implemented agent, determining, by the computer-implemented agent using the utterance, two or more second computer-implemented agents each of which can respond to the utterance;
in response to determining the two or more second computer-implemented agents each of which can respond to the utterance, providing a prompt that requests selection of one of the second computer-implemented agents each of which can respond to the utterance;
in response to providing the prompt that requests selection of one of the second computer-implemented agents each of which can respond to the utterance, receiving an indication of a selected one of the second computer-implemented agents; and
determining, by the computer-implemented agent using the received indication of the selected one of the second computer-implemented agents, that the utterance specifies a request for the selected one of the second computer-implemented agents;

providing, by the computer-implemented agent to the other computer-implemented agent, a representation of the utterance and data identifying the sentiment and user profile data for a user who spoke the utterance to enable the other computer-implemented agent to automatically complete a transaction using the user profile data without requesting, from the user, data that is included in the user profile data;
receiving, by the user device, a response to the utterance for the other computer-implemented agent that was determin

Assignees

• Google Llc

Inventors

• Chen Johnny
• Dean Thomas L
• Lau Qiangfeng Peter
• Gandhe Sudeep
• Schine Gabriel

Classifications

• G10L13/08

  Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G06F21/6245

  Protecting personal data, e.g. for financial or medical purposes · CPC title

• G10L15/22 Primary

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

• G06F17/30654

  Physics · mapped topic