Virtual Agent Emotion and Context Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtual agents lack contextual understanding and relatability, relying on limited information for responses and lacking a visually appealing and emotionally engaging presence, which limits their interaction with users.
Innovation Solution
A method for operating a virtual agent that involves obtaining an interaction context, determining an agent state, updating emotion and mood based on this context and personality, and selecting behaviors that adjust its appearance or generate audio, allowing for a more human-like interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If virtual agents use limited information for responses, then processing speed is maintained, but contextual understanding and relatability deteriorate
Solution Approach 1:
The patent segments information processing into multiple contextual layers (communication context, target information, situation information, people information) that are processed independently and then integrated. This allows comprehensive contextual understanding without overwhelming the processing system with undifferentiated data.
Solution Approach 2:
The patent implements preliminary action by pre-defining context types and response frameworks before interactions occur. The system prepares multiple possible responses and selects/appropriates them based on contextual analysis, rather than generating responses from scratch during each interaction.
2Adaptability or versatility
If virtual agents lack simulated personality and emotions, then computational resources are conserved, but user engagement and naturalness of interaction deteriorate
Solution Approach 1:
The patent applies partial action by implementing emotion simulation at appropriate levels - not all aspects of human emotion are simulated, but key emotional responses relevant to the interaction context are generated. This provides sufficient naturalness without requiring full computational overhead of complete emotional modeling.
Solution Approach 2:
The system uses parameter changes to represent emotional states through adjustments in response tone, timing, and content selection rather than complex internal emotional modeling. Emotional adaptability is achieved by modifying output parameters based on detected context and predefined personality traits.
3Adaptability or versatility
If virtual agents lack visual expressions, then system complexity is reduced, but visual appeal and emotional engagement deteriorate
Solution Approach 1:
The patent uses copying by implementing visual expressions that mirror or correspond to the simulated emotional states rather than creating entirely new visual systems. The avatar's visual representation copies human emotional cues (facial expressions, body language) that are generated through the emotion simulation framework.
Solution Approach 2:
The visual expression system serves multiple functions simultaneously - it conveys emotional state, enhances user engagement, and provides visual feedback about system status. A single visual expression framework accomplishes what would otherwise require separate systems for each function.
4Productivity
If virtual agents rely on simple response mechanisms, then response time is reduced, but naturalness and contextual appropriateness of dialogue deteriorate
Solution Approach 1:
The system prepares multiple possible responses in advance for different contextual scenarios. During actual interaction, the system selects from these pre-prepared responses and makes appropriate modifications, rather than generating responses from scratch. This maintains speed while improving naturalness.
Solution Approach 2:
The patent uses parameter changes to adapt predefined responses to specific contexts - modifying tone, timing, content emphasis, and other parameters of base responses to make them contextually appropriate without requiring complete response regeneration.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
In one aspect, a method for operating a virtual agent will be described. An interaction context is obtained. An agent state is determined based on the obtained interaction context. The agent state indicates an activity of the virtual agent. The emotion of the virtual agent is updated based on the obtained interaction context, the determined agent state and/or a personality of the virtual agent. One or more behaviors are selected. Each behavior involves or indicates a change in an appearance of the virtual agent or generation of audio. Various embodiments relate to devices, servers, software and systems arranged to implement one or more of the above operations.