Virtual-Space Communication Assistance with Gaze-Speech Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems require manual group setting operations during remote meetings, necessitating user intervention to form and change groups, disrupting conversations.
Innovation Solution
A communication assistance system that automatically identifies conversation partners based on line of sight and speech detection, adjusting audio volumes and display settings to facilitate smoother group interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual group setting operation is used, then users can form groups for conversation, but users need to manually operate to change groups during conversation, disrupting the flow
Solution Approach 1:
The system automatically performs group formation and switching based on detected speech and line-of-sight data, eliminating the need for manual user operations. The system serves itself by autonomously determining which users should be in the same group and automatically switching groups when conversation partners change, thereby resolving the contradiction between ease of operation and time loss.
Solution Approach 2:
The system performs preliminary group formation based on detected speech and line-of-sight information before users need to switch groups. By proactively identifying conversation partners and pre-configuring groups, the system eliminates the need for manual intervention during conversation, reducing time loss while maintaining ease of operation.
2Productivity
If automatic group formation is implemented, then group switching becomes seamless, but system complexity increases
Solution Approach 1:
The system segments the automatic group formation process into distinct functional modules: speech detection module, line-of-sight detection module, conversation partner identification module, and group management module. Each module handles a specific aspect of the automatic process, reducing overall system complexity while maintaining high conversation efficiency through coordinated operation of these segmented components.
Solution Approach 2:
The system introduces an intermediary control module that mediates between the detection modules (speech and line-of-sight) and the group management functions. This intermediary layer processes raw detection data, identifies conversation partners, and translates this information into group configuration commands, thereby simplifying the overall system architecture while enabling seamless group switching for improved productivity.
Data Source
AI summary
A communication assistance system related to one embodiment assists audio communication among a plurality of users. The system includes at least one processor that: specifies a second user who is a conversation partner of a first user, based on a positional relationship between a line of sight of the first user with respect to a screen and user objects, and a detection result of a speech of the first user, the screen showing a virtual space in which the user objects respectively corresponding to the users are arranged, the virtual space being two-dimensional or three-dimensional; sets a group including the first user and the second user; and outputs, to the first user, a speech voice of the second user who belongs to the group with a larger volume than a speech voice of the second user in a case where the second user does not belong to the group.