- Number
- 20200320427
- Published
- 2020-10-08
- Filed
- 2019-04-03
- Assignee
- Disney Enterprises, Inc.
- Inventors
- Kennedy; James R., Lehman; Jill Fain, El Haddad; Kevin, Fidaleo; Douglas A.
- CPC
- G06N3/09; G06N3/0464; G06N3/006; G06F16/24573; G06N20/00; G06N3/08
- Verdict
- Low Notable software
- Source
- Google Patents · FreePatentsOnline
The keeper's note
ML-driven expression generation for an interactive social agent/character.
Abstract
A system providing an interactive social agent can include a computing platform having a hardware processor and a memory storing a training content standardization software code configured to receive content depicting human expressions and including annotation data describing the human expressions from multiple content annotation sources, generate a corresponding content descriptor for each content annotation source to translate the annotation data into a standardized data format, and transform the annotation data into the standardized data format using the corresponding content descriptor. The content and the annotation data in the to standardized format are stored as training data for use in training expressions for the interactive social agent. The memory may also store a character remapping software code configured to receive data identifying an expression for the interactive social agent, identify a character persona of the interactive social agent, and determine a modified expression based on expressive idiosyncrasies of the character persona.
Background
BACKGROUND
Advances in artificial intelligence have led to the development of a variety of devices providing one of several dialogue-based interfaces, such as GOOGLE HOME™, AMAZON ALEXA™, and others. However, the dialogue interfaces to provided by these and similar devices tend to lack character and naturalness. Moreover, these devices and the dialog interfaces provided by the conventional art are typically transactional, and indicate to a user that they are listening for a communication from the user by responding to an affirmative request by the user. In contrast to such transactional interactions, natural communications between human beings are more nuanced and include the use of nonverbal, as well as verbal expressions. Consequently, there is a need in the art for an automated solution capable of interpreting contextual inputs in order to determine naturalistic expressions for use by an interactive social agent in communicating with a user.SUMMARY
There are provided systems and methods for performing automated determination of expressions for an interactive social agent, substantially as shown in and/or described in connection with at least one of the figures, and as set forth more completely in the claims.
Claims
1. A system providing an interactive social agent, the system comprising: a computing platform including a hardware processor and a system memory; a training content standardization software code stored in the system memory; the hardware processor configured to execute the training content standardization software code to: receive a content depicting human expressions and including annotation data describing the human expressions from each of a plurality of content annotation sources; for each of the plurality of content annotation sources, generate a corresponding content descriptor configured to translate the annotation data into a standardized data format; for each of the plurality of content annotation sources, transform the annotation data into the standardized data format using the corresponding content descriptor; and for all of the plurality of content annotation sources, store the content and the annotation data in the standardized format as a training data in a training database for use in training an expression by the interactive social agent in communicating with a user.
7. A method for use by a system providing an interactive social agent, the system including a computing platform having a hardware processor and a system memory storing a training content standardization software code, the method comprising: receiving, by the training content standardization software code executed by the hardware processor, a content depicting human expressions and including annotation data describing the human expressions from each of a plurality of content annotation sources; for each of the plurality of content annotation sources, generating, by the training content standardization software code executed by the hardware processor, a corresponding content descriptor configured to translate the annotation data into a standardized data format; for each of the plurality of content annotation sources, transforming, by the training content standardization software code executed by the hardware processor, the annotation data into the standardized data format using the corresponding content descriptor; and for all of the plurality of content annotation sources, storing, by the training content standardization software code executed by the hardware processor, the content and the annotation data in the standardized format as a training data in a training database for use in training an expression for use by the interactive social agent in communicating with a user.
13. A system providing an interactive social agent, the system comprising: a computing platform including a hardware processor and a system memory; a character remapping software code stored in the system memory; the hardware processor configured to execute the character remapping software code to: receive a data identifying an expression for use by the interactive social agent in communicating with a user; identify a character persona assumed by the interactive social agent; obtain a personality profile of the character persona, the personality profile including at least one expressive idiosyncrasy of the character persona; determine a modified expression specifically for the character persona assumed by the interactive social agent based on the data identifying the expression and the expressive idiosyncrasy of the character persona; and render the modified expression using the interactive social agent.
17. A method for use by a system providing an interactive social agent, the system including a computing platform having a hardware processor and a system memory storing a character remapping software code, the method comprising: receiving, by the character remapping software code executed by the hardware processor, a data identifying an expression for use by the interactive social agent in communicating with a user; identifying, by the character remapping software code executed by the hardware processor, a character persona assumed by the interactive social agent; obtaining, by the character remapping software code executed by the hardware processor, a personality profile of the character persona, the personality profile including at least one expressive idiosyncrasy of the character persona; determining, by the character remapping software code executed by the hardware processor, a modified expression specifically for the character persona assumed by the interactive agent based on the data identifying the expression and the expressive idiosyncrasy of the character persona; and rendering, by the character remapping software code executed by the hardware processor, the modified expression using the interactive social agent.