Video Telephone Speaker Recognition and Dynamic Background Synthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video telephone systems used by multiple persons, such as families, face usability issues in identifying transmission or reception counterparts and selecting suitable backgrounds, as existing methods are insufficient for diverse usage environments and situations.
Innovation Solution
A video telephone device and method that includes an imaging unit, image processing unit, communication unit, display unit, individual information storage unit, and speaker recognition unit, allowing for image synthesis and background selection based on recognized speakers and settings, enabling clear identification and communication between users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single account is used for video telephone service in family devices, then device complexity is reduced, but it becomes difficult to identify transmission or reception counterparts and select appropriate backgrounds
Solution Approach 1:
The patent segments the single family account into multiple member accounts, each with unique identification information. This allows the system to distinguish between different family members using the device, enabling proper counterpart identification and background selection while maintaining a unified family account structure at the service level
Solution Approach 2:
The patent creates virtual copies of user profiles and settings for each family member within the shared account. Each member has their own identification information, speaker recognition data, and background image settings stored as separate records under the family account, allowing personalized operation without requiring separate physical devices or complex account structures
2Adaptability or versatility
If multiple members are registered with the same account, then usability for family use is improved, but it becomes difficult to determine who the transmission or reception counterpart is
Solution Approach 1:
The patent assigns unique local quality attributes (identification information) to each member within the account. Each member has distinct speaker recognition information and is associated with specific background images, allowing the system to identify which member is using the device and select appropriate settings for that specific user while maintaining shared account access
3Extent of automation
If automatic background selection based on telephone number is implemented, then background selection is simplified, but it is insufficient for diverse usage environments and situations in multi-user devices
Solution Approach 1:
The patent makes the background selection system dynamic by associating different background images with different members and their respective usage situations. The system automatically selects backgrounds based on both the telephone number and the identified member's preferences, allowing each user to have personalized background settings that adapt to different call scenarios while maintaining automated selection
Solution Approach 2:
The patent creates a universal background management system that works across multiple members and usage scenarios. The same automated background selection mechanism serves all family members, but each member has their own set of background images and preferences, allowing the system to handle diverse usage environments through a unified multi-functional approach
Data Source
AI summary
Provided are a video telephone device and a video telephone processing method in which a video call environment with excellent usability is provided even when used by a plurality of persons. A video telephone device that performs a call with another video telephone device via a network photographs an image of a speaker, recognizes the speaker based on the photographed image, and changes a background image in an image to be transmitted according to a recognition result. Thus, for example, when the users who are not close to each other perform a call, a situation in which an image showing a state of a room is supplied as a background image is solved.