Multi-party Conversational Agent with Dynamic Dialogue Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional conversational agents are unable to engage in natural, fluid conversations with multiple users simultaneously and lack emotional sophistication due to their transactional design, which limits user personalization by only remembering a few past interactions and predefined keywords.
Innovation Solution
The development of multi-party conversational agents that utilize a computing platform with an input module for sensing user intents, an output module for expressive interaction, and software code that includes a dialogue manager and user history database to coordinate conversations, allowing for dynamic switching between group and dyadic dialogues and remembering user preferences for personalized interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional conversational agents use transactional design with limited memory, then device complexity is reduced, but the ability to engage in natural multi-party conversations and provide emotional sophistication deteriorates
Solution Approach 1:
The system segments the conversation management into multiple independent components: conversation queues for different user groups, interruption stacks for handling simultaneous inputs, and separate memory structures for different users. This allows the system to handle multiple parties independently while maintaining overall coordination, resolving the contradiction between simple architecture and multi-party capability.
Solution Approach 2:
The patent introduces temporal and hierarchical dimensions to conversation management. Conversation queues organize interactions chronologically, while interruption stacks add a priority layer. This dimensional approach enables the system to manage complex multi-party interactions without requiring exponential increase in base system complexity.
2Device complexity
If conversational agents remember only a few past interactions and predefined keywords, then memory requirements and processing complexity are minimized, but user personalization and emotional sophistication are limited
Solution Approach 1:
The system implements local quality by maintaining separate, specialized memory structures for different users rather than a single generic memory. Each user has their own conversation history and preference profile, allowing personalized interactions without requiring the system to process and store all possible user data centrally, thus managing complexity while enhancing personalization.
Solution Approach 2:
The system performs preliminary actions by pre-processing and structuring user preferences and conversation histories into organized formats before interactions occur. User profiles and conversation contexts are prepared in advance, enabling quick retrieval and personalized responses without requiring complex real-time processing during actual conversations.
3Adaptability or versatility
If conversational agents handle multiple users simultaneously, then user engagement and application value are enhanced, but coordination complexity and response time management increase
Solution Approach 1:
The patent introduces conversation queues as intermediary structures that buffer and organize inputs from multiple users. These queues act as mediators between users and the processing system, allowing concurrent interactions to be managed systematically. The interruption stack serves as another intermediary layer that prioritizes and coordinates responses, reducing direct coordination complexity while enabling multi-user engagement.
4Speed
If conversational agents use simple transactional responses, then response generation speed is maintained, but emotional sophistication and natural conversation flow are reduced
Solution Approach 1:
The system implements dynamics by adapting response generation based on conversation context, user preferences, and interaction history. Rather than using fixed transactional responses, the system dynamically selects and customizes responses based on real-time conversation state and stored user profiles, enabling emotional sophistication while maintaining efficient response generation through pre-structured response templates.
Data Source
AI summary
A multi-party conversational agent includes a computing platform having a hardware processor and a memory storing a software code. The hardware processor is configured to execute the software code to identify a first predetermined expression for conversing with a group of people, and to have a group conversation, using the first predetermined expression, with at least some members of the group. The hardware processor is configured to further execute the software code to identify, while having the group conversation, a second predetermined expression for having a dialogue with at least one member of the group, and to interrupt the group conversation to have the dialogue, using the second predetermined expression, with the at least one member of the group.