Environment-responsive user interface/entertainment device that simulates personal interaction

• US 6,721,706 B1
• Filed: 10/30/2000
• Issued: 04/13/2004
• Est. Priority Date: 10/30/2000
• Status: Expired due to Term

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Gutta, Srinivas, Pelletier, Daniel, Eshelman, Larry J., Strubbe, Hugo J., Milanski, John
• Primary Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US09/699,606
• Time in Patent Office: 1,261 Days
• Field of Search: 704/200, 704/246, 704/251, 704/270, 704/271, 704/257, 128/904
• US Class Current: 704/275
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/228 of application context

Abstract

An interaction simulator, such as a chatterbot, is enabled to simulate an awareness of the user to generate an interaction that is more natural and appropriate than prior art chatterbots. For example, the device may employ machine vision to detect the number of persons present or the activity of the user and respond accordingly by interrupting its output or by inviting a conversation or other interaction when a user approaches. The device may modify its responses according to the user'"'"'s activity, for example, by playing music when the user falls asleep or requesting an introduction when another user speaks. The device may also respond to unrecognized changes in the situation by inquiring about what is going on to stimulate interaction or generate new responses.

Claims

1. A conversation simulator for simulating conversational interaction with a user, comprising:

a controller programmed to receive first, second, and third user inputs, each chronologically separated by at least one reply from said controller;

said first, second, and third user inputs each including a natural language declaration as distinguished from a query or a command;

an output device connected to said controller to output said corresponding replies;

a first input device connected to apply said first, second, and third user inputs to said controller;

at least one second input device to receive environmental data indicating an environment of said user;

said controller being programmed to classify said environment of said user responsively to said environmental data to produce class data, wherein said class data includes an indicator of a physical location of said user; and

said controller being programmed to generate said at least one reply responsively to a corresponding one of said first, second and third user inputs and said class data.

2. A device as in claim 1, wherein said environmental data includes at least one of image and audio data.

3. A device as in claim 1, wherein said class data includes an indicator of a body position of said user.

4. A device as in claim 1, wherein said class data includes data indicating a presence of a person other than said user.

5. A device as in claim 1, wherein said class data includes data indicating an activity of said user.

6. A device as in claim 1, wherein said second input device includes at least one of a video camera and an audio detector.

7. A method of generating a dialogue response in a conversation simulator, comprising the steps of:

receiving a natural language declaration from a user;

determining a condition of an environment of said user including an indication of a physical location of said user;

generating a response to said declaration responsively to a result of said step of determining and to a content of said declaration.

8. A method as in claim 7, wherein said step of receiving includes inputting speech through an audio transducer.

10. A method as in claim 7, wherein said step of determining includes classifying image data from a camera, classifying audio data from an audio transducer.

11. A method as in claim 7, wherein said step of determining includes:

classifying image data from a camera;

classifying audio data from an audio transducer;

combining results of said first and second steps of classifying to determine said condition.

12. A method as in claim 11, wherein said step of receiving includes inputting speech through an audio transducer.

13. A method as in claims 11, wherein said step of determining includes classifying at least one of image and audio data.

14. A method as in claim 11, wherein said step of determining includes classifying input data and determining a physical location of said user responsively to said input data.

15. A method as in claim 11, wherein said step of determining includes classifying input data and determining an activity of said user responsively to said input data.

9. A method as in claim wherein 7, said step of determining includes classifying image data from a camera.

16. A method of generating a dialogue response in a conversation simulator, comprising the steps of:

receiving input from a user;

determining a condition of an environment of said user including data indicating a presence of a person other than said user;

generating a conditional response responsively to a result of said step of determining and to a content of said input, a condition of said conditional response being said result of said step of determining.

17. A device as in claim 16, wherein said step of determining includes classifying at least one of image and audio data.

18. A device as in claim 16, wherein said step of determining includes classifying input data and determining a physical location of said user responsively to said input data.

19. A device as in claim 16, wherein said step of determining includes classifying input data and determining an activity of said user responsively to said input data.