Dialog simulation

• US 10,272,349 B2
• Filed: 09/07/2016
• Issued: 04/30/2019
• Est. Priority Date: 09/07/2016
• Status: Active Grant

• Current Assignee: Isaac Davenport
• Original Assignee: Isaac Davenport
• Inventors: Davenport, Isaac
• Primary Examiner(s): Le, Thuykhanh
• Application Number:
  US15/258,609
• Publication Number:
  US 20180065054A1
• Time in Patent Office: 965 Days
• Field of Search: None
• CPC Class Codes: A63H 2200/00 Computerized interactive toys, e.g. dolls
• CPC Class Codes: A63H 3/28 Arrangements of sound-producing means in dolls; Means in dolls for producing sounds
• CPC Class Codes: A63H 5/00 Musical or noise- producing devices for additional toy effects other than acoustical musical tops A63H1/28; sound-producing dolls A63H3/28; musical toy vehicles A63H17/268; acoustical vehicle-signalling devices A63H17/32; whistling toy locomotive arrangements A63H19/14; music, sound or noise production in general G10
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation
• CPC Class Codes: G06N 3/008 based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour

Abstract

A method of producing simulated dialog outputs selected phrases in a sequence defined by a selected dialog model. When a state change is recorded one or more of dialog models which have non-zero state change type weights corresponding to the recorded state change type are identified and one of the identified dialog models is selected probabilistically with influence from the non-zero state change type weights. One or more unique characters are mapped to character indices of the selected dialog model and one or more phrases defined in the one or more mapped unique characters are identified which have non-zero phrase type weights corresponding to the phrase types of the selected dialog model. For each phrase type defined in the selected dialog model, at least one of the phrases are selected probabilistically with influence from the non-zero phrase type weights.

Claims

1. A method of producing simulated dialog, comprising:

defining each of a plurality of unique characters in terms of a plurality of phrases, wherein each of the plurality of phrases is weighted according to one or more phrase type weights;

defining each of a plurality of dialog models in terms of state change type weights, character indices and phrase types;

in a memory, associating each of the plurality of unique characters with one of a plurality of unique identifier codes, wherein each of the plurality of unique identifier codes is broadcast by one of a plurality of transceivers, wherein each of the plurality of transceivers is coupled with one of a plurality of toys;

recording a state change type of the plurality of transceivers;

identifying one or more of the dialog models which have non-zero state change type weights corresponding to the recorded state change type;

selecting one of the identified dialog models probabilistically with influence from the non-zero state change type weights;

associating each of the plurality of toys with one of the plurality of unique characters;

mapping one or more of the unique of the characters to the character indices of the selected dialog model;

for each of the plurality of mapped unique characters, identifying one or more phrases defined therein which have non-zero phrase type weights corresponding to the phrase types of the selected dialog model;

for the phrase types defined in the selected dialog model, selecting, probabilistically with influence from the non-zero phrase type weights, a plurality of the identified phrases; and

outputting the selected phrases in a sequence defined by the selected dialog model.

2. The method as set forth in claim 1, wherein defining each of a plurality of dialog models in terms of state change type weights, character indices and phrase types further comprises defining each of the plurality of dialog models in terms of motion type weights;

and

wherein recording a state change type further comprises recording a motion type.

3. The method as set forth in claim 2, further comprising providing a branching storyline by linking two or more of the dialog models together in a sequence in response to two or more consecutively recorded motion types corresponding with the linked dialog models.

4. The method as set forth in claim 2, wherein identifying one or more dialog models with non-zero state change type weights corresponding to the recorded state change type further comprises identifying one or more dialog models with non-zero motion type weights corresponding to the recorded motion type.

5. The method as set forth in claim 1, wherein recording the state change type further comprises monitoring input from the plurality of transceivers or from sensors coupled with one or more of the plurality of toys.

6. The method as set forth in claim 1, wherein outputting the selected phrases in the sequence further comprises outputting an audible sequence of pre-recorded phrases.

7. The method as set forth in claim 1, wherein defining each of a plurality of unique characters in terms of a plurality of phrases each weighted according to one or more phrase type weights further comprises defining the plurality of unique characters in terms of descriptive metadata.

8. The method as set forth in claim 1, further comprising transmitting, over a communications network, definitions of the defined plurality of unique characters including the plurality of phrases and one or more phrase type weights.

9. The method as set forth in claim 1, further comprising transmitting, over a communications network, definitions of the defined dialog models including the state change type weights, character indices and phrase types.

10. A system for producing simulated dialog, comprising:

a memory having recorded thereto;

a dialog simulator comprising;

computer-readable definitions of each of a plurality of unique characters in terms of a plurality of phrases, wherein each of the plurality of phrases is weighted according to one or more phrase type weights,

computer-readable definitions of each of a plurality of dialog models in terms of state change type weights, character indices and phrase types, and

a dialog engine including computer readable instructions for producing a simulated dialog;

a processor configured to, according to computer readable instructions of the dialog engine;

in the memory, associate each of the plurality of unique characters with one of a plurality of unique identifiers, wherein each of the plurality of unique identifiers is broadcast by one of a plurality of transceivers, wherein each of the plurality of transceivers is coupled with one of a plurality of toys,

record a state change type of the plurality of transceivers,

identify one or more of the dialog models which have non-zero state change type weights corresponding to the recorded state change type of the simulator,

select one of the identified dialog models probabilistically with influence from the non-zero state change type weights,

associate each of the plurality of toys with one of the plurality of unique characters, map one or more of the unique characters to the character indices of the selected dialog model,

for each of the plurality of mapped unique characters, identify one or more phrases defined therein which have non-zero phrase type weights corresponding to the phrase types of the selected dialog model, and

for the phrase types defined in the selected dialog model, select probabilistically with influence from the non-zero phrase type weights, a plurality of the identified phrases; and

an output configured to present the selected phrases in a sequence defined by the selected dialog model.

11. The system as set forth in claim 10, wherein:

the state change type weights further comprise motion type weights; and

in order to record a state change type, the processor is further configured to record a motion type associated with one or more of the plurality of unique characters.

12. The system as set forth in claim 11, wherein, to provide a branching storyline, the processor is further configured to link two or more of the dialog models together in a sequence in response to two or more consecutively recorded motion types corresponding with the linked dialog models.

13. The system as set forth in claim 11, wherein, in order to identify one or more dialog models with non-zero state change type weights corresponding to the recorded state change type, the processor is further configured to identify one or more dialog models with non-zero motion type weights corresponding to the recorded motion type.

14. The system as set forth in claim 10, wherein, in order to record a state change type, the processor is further configured to monitor input from the plurality of transceivers or from sensors coupled with one or more of the plurality of toys.

15. The system as set forth in claim 10, wherein the phrases are pre-recorded and the output is an audio speaker configured to present the pre-recorded phrases audibly, in the sequence.

16. The system as set forth in claim 10, wherein the definitions of each of the plurality of unique characters further comprise descriptive metadata.

17. The system as set forth in claim 10, further comprising a dialog server configured to transmit to the memory, over a communications network, the definitions of each of the plurality of unique characters including the plurality of phrases and one or more phrase type weights.

18. The system as set forth in claim 10, further comprising a dialog server configured to transmit to the memory, over a communications network, the definitions of the plurality of dialog models including the state change type weights, character indices and phrase types.