Emotionally responsive artificial intelligence interactive character
A system includes a computing platform having a hardware processor and a memory storing software code, a memory data structure storing memory features for an artificial intelligence interactive character (AIIC), and a trained machine learning (ML) model. The hardware processor executes the software code to receive interaction data describing a communication by a user with the AIIC, predict, using the trained ML model and the interaction data, at least one user memory feature(s) of the communication, and identify, using the memory data structure, one or more of the memory features for the AIIC as corresponding to the user memory feature(s). The software code also determines, using the user memory feature(s) and the corresponding one or more of the memory features for the AIIC, an interactive communication for execution by the AIIC in response to the communication by the user; and outputs the interactive communication to the AIIC.
1 . A system comprising:
a computing platform having a hardware processor and a system memory;
the system memory storing a software code, a memory data structure storing a plurality of memory features for an artificial intelligence interactive character (AIIC), and a trained machine learning (ML) model;
the hardware processor configured to execute the software code to:
receive interaction data describing a communication by a user with the AIIC;
predict, using the trained ML model and the interaction data, at least one user memory feature of the communication;
identify, using the memory data structure, one or more of the plurality of memory features for the AIIC as corresponding to the at least one user memory feature, wherein identifying comprises comparing the at least one user memory feature and the one or more of the plurality of memory features across multiple dimensions, and computing an aggregated similarity score using dimension weights;
determine, using the at least one user memory feature of the communication and the corresponding one or more of the plurality of memory features for the AIIC, an interactive communication for execution by the AIIC in response to the communication by the user; and
output the interactive communication to the AIIC.
2 . The system of claim 1 , wherein the plurality of memory features stored by the memory data structure comprise at least one of human generated memory features or synthesized memory features for the AIIC.
3 . The system of claim 1 , wherein determining the interactive communication for execution by the AIIC in response to the communication by the user comprises evaluating at least one candidate interactive communication output by a large-language ML model accessible by the software code.
4 . The system of claim 1 , wherein the hardware processor is further configured to execute the software code to:
control the AIIC to execute the interactive communication.
5 . The system of claim 1 , wherein the AIIC comprises a virtual character, and wherein the interactive communication output to the AIIC comprises at least one of speech, a gesture, a facial expression, or a posture for execution by the virtual character.
6 . The system of claim 1 , wherein the AIIC comprises a machine, and wherein the interactive communication output to the AIIC comprises at least one of speech, movement, a facial expression, or a posture for execution by the machine.
7 . The system of claim 1 , wherein the memory data structure comprises one of an undirected cyclic graph or an acyclic graph.
8 . The system of claim 1 , wherein the predicted at least one user memory feature comprises a plurality of predicted user memory features, and wherein the hardware processor is further configured to execute the software code to:
represent the plurality of predicted user memory features as a knowledge graph having a same data structure as the memory data structure; and
wherein identifying the one or more of the plurality of memory features for the AIIC as corresponding to the at least one user memory feature further uses the knowledge graph.
9 . The system of claim 1 , wherein the communication by the user with the AIIC comprises at least one of speech, text, a non-verbal vocalization, or a gesture by the user, or a facial expression or a posture by the user.
10 . The system of claim 1 , wherein the communication by the user with the AIIC comprises speech by the user, and wherein the interaction data describes a prosody of the speech.
11 . A method for use by a system including a computing platform having a hardware processor and a system memory, the system memory storing a software code, a memory data structure storing a plurality of memory features for an artificial intelligence interactive character (AIIC), and a trained machine learning (ML) model, the method comprising:
receiving, by the software code executed by the hardware processor, interaction data describing a communication by a user with the AIIC;
predicting, by the software code executed by the hardware processor and using the trained ML model and the interaction data, at least one user memory feature of the communication;
identifying, by the software code executed by the hardware processor and using the memory data structure, one or more of the plurality of memory features for the AIIC as corresponding to the at least one user memory feature, wherein identifying comprises comparing the at least one user memory feature and the one or more of the plurality of memory features across multiple dimensions, and computing an aggregated similarity score using dimension weights;
determining, by the software code executed by the hardware processor, using the at least one user memory feature of the communication and the corresponding one or more of the plurality of memory features for the AIIC, an interactive communication for execution by the AIIC in response to the communication by the user; and
outputting, by the software code executed by the hardware processor, the interactive communication to the AIIC.
12 . The method of claim 11 , wherein the plurality of memory features stored by the memory data structure comprise at least one of human generated memory features or synthesized memory features for the AIIC.
13 . The method of claim 11 , wherein determining the interactive communication for execution by the AIIC in response to the communication by the user comprises evaluating at least one candidate interactive communication output by a large-language ML model accessible by the software code.
14 . The method of claim 11 , further comprising:
controlling the AIIC, by the software code executed by the hardware processor, to execute the interactive communication.
15 . The method of claim 11 ,
wherein the AIIC comprises a virtual character,
wherein the interactive communication output to the AIIC comprises at least one of speech, a gesture, a facial expression, or a posture for execution by the virtual character, and
wherein the method further comprises controlling the virtual character, by the software code executed by the hardware processor, to execute the interactive communication.
16 . The method of claim 11 ,
wherein the AIIC comprises a machine,
wherein the interactive communication output to the AIIC comprises at least one of speech, movement, a facial expression, or a posture for execution by the machine, and
wherein the method further comprises controlling the machine, by the software code executed by the hardware processor, to execute the interactive communication.
17 . The method of claim 11 , wherein the predicted at least one user memory feature comprises a plurality of predicted user memory features, the method further comprising:
representing, by the software code executed by the hardware processor, the plurality of predicted user memory features as a knowledge graph having a same data structure as the memory data structure; and
wherein identifying the one or more of the plurality of memory features for the AIIC as corresponding to the at least one user memory feature further uses the knowledge graph.
18 . The method of claim 11 , wherein the communication by the user with the AIIC comprises at least one of speech, text, a non-verbal vocalization, or a gesture by the user, or a facial expression or a posture by the user.
19 . The method of claim 11 , wherein the communication by the user with the AIIC comprises speech by the user, and wherein the interaction data describes a prosody of the speech.
20 . The method of claim 16 , wherein the machine comprises a robot.