Virtual Assistant With Real-Time Emotions

• US 20080096533A1
• Filed: 12/28/2006
• Published: 04/24/2008
• Est. Priority Date: 10/24/2006
• Status: Abandoned Application

• Current Assignee: Kallideas SPA
• Original Assignee: Kallideas SPA
• Inventors: Gribaudo, Claudio, Manfredi, Giorgio
• Application Number:
  US11/617,150
• Publication Number:
  US 20080096533A1
• Time in Patent Office: Days
• US Class Current: 455/412.1
• CPC Class Codes: G06N 3/006 based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]

Abstract

A modular digital assistant that detects user emotion and modifies its behavior accordingly. The desired emotion is produced in a first module and a transforming module then converts the emotion into the desired output medium. The degree or subtleness of the emotion can be varied. Where the emotion is not completely clear, the virtual assistant may prompt the user. The detected emotion can be used for the commercial purposes the virtual assistant is helping the user with. Various primary emotional input indicators are combined to determine a more complex emotion or secondary emotional state. The user'"'"'s past interactions are combined with current emotion inputs to determine a users emotional state.

Claims

1. A virtual assistant comprising:

a user input for providing information about a user emotion;

an input transform module for transforming said information into normalized emotion data; and

a core module for producing a virtual assistant emotion for the virtual assistant based on detected user emotion.

2. The virtual assistant of claim 1 further comprising an adjustment module configured to apply an adjustment to the degree of said virtual assistant emotion based on a context.

3. The virtual assistant of claim 2 wherein said context comprises one of a user profile and a type of service provided by said virtual assistant.

4. The virtual assistant of claim 1 further comprising an output transform module configured to transform a normalized emotion output for said virtual assistant into one of a voice rendering, a video, and a text message.

5. The virtual assistant of claim 1 further comprising:

a right brain module configured to determine a probability of veracity of said user emotion;

said right brain module being further configured to compare said probability to a threshold; and

said right brain module being further configured to formulate a stimulus to provide more data to determine said user emotion if said probability is below said threshold.

6. A virtual assistant comprising:

a user input for providing information about a user emotion;

a connecting layer configured to provide an output emotion prior to calculating a response to a user;

an artificial intelligence engine configured to calculate a response to a user input.

7. A virtual assistant comprising:

a user input device for providing input information from a user;

an emotion detection module configured to detect a user'"'"'s emotion from said input information;

a core module for producing a virtual assistant emotion for the virtual assistant based on said user'"'"'s emotion.

8. The virtual assistant of claim 7 wherein said input information is an image and said user'"'"'s emotion is detected from one of a facial expression of said user and a gesture of said user.

9. A virtual assistant comprising:

a first media input from a user;

a second media input from said user;

an emotion detection module configured to detect said user'"'"'s emotion from a combination of said media inputs;

a core module for producing a virtual assistant emotion for the virtual assistant based on said user'"'"'s emotion.

10. The virtual assistant of claim 9 wherein

said first media input is one of a voice and text input;

and

said second media input is a camera input.

11. The virtual assistant of claim 9 wherein said emotion module is further configured to consult, in determining said user'"'"'s emotion, one of a user profile and group characteristics of a group said user is associated with.

12. An user help system comprising:

a user input for providing a user dialogue and user emotion information; and

an expert system for providing a response to said user dialogue, wherein said response varies based on said user emotion information.

13. The system of claim 12 wherein said response varies in one of a price and an alternative option.

14. A method for controlling a virtual assistant comprising:

receiving a user input;

analyzing said user input to detect at least one user emotion;

producing a virtual assistant emotion for the virtual assistant based on said detected user emotion;

said virtual assistant emotion also being produced based on one of a user profile and a type of service provided by said virtual assistant.

15. The method of claim 14 further comprising applying an adjustment to the degree of said virtual assistant emotion based on a context.

16. The method of claim 14 further comprising:

transforming said user emotion into normalized emotion data; and

transforming said virtual assistant emotion into a media specific virtual assistant emotion.

17. The method of claim 14 further comprising:

offering said user an accommodation in response to detection of a predetermined emotion above a predetermined level.

18. The method of claim 17 wherein said accommodation is a discount.

19. The method of claim 14 further comprising:

detecting an ambiguous user emotion; and

forming a virtual assistant question, unrelated to a current dialogue with said user, to elicit more information on an emotion of said user.