Comment Voice Synthesis With Character Overlay for Live Streaming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing comment reading technologies in live streaming, such as mechanical voices, can be monotonous and lead to viewer disengagement due to boredom, and distributors' manual reading may result in comments being skipped.
Innovation Solution
A content generation device that synthesizes voices from comments and generates character content to superimpose on the content, allowing for dynamic and engaging live streaming experiences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a distributor reads comments manually, then the interaction feels natural and authentic, but comments may be skipped and viewers lose interest
Solution Approach 1:
The system uses automated voice synthesis to read comments without requiring the distributor's manual intervention, allowing continuous and complete comment reading while maintaining natural-sounding delivery through synthesized voices
Solution Approach 2:
The patent replaces the mechanical action of manual reading with automated voice synthesis technology, enabling comments to be read aloud automatically while maintaining engagement through varied and natural-sounding synthesized voices
2Reliability
If mechanical voice synthesis is used to read comments, then comment skipping is avoided, but the monotonous synthesized voice makes viewers bored
Solution Approach 1:
The system changes the parameters of voice synthesis to create varied and natural-sounding voices instead of monotonous ones, adjusting voice characteristics such as tone, pace, and emotion to maintain viewer engagement while ensuring complete comment reading
Solution Approach 2:
The patent introduces dynamic voice synthesis that adapts to different comment contexts and speakers, creating varied and engaging voice patterns rather than static monotonous repetition, thereby maintaining viewer interest throughout the broadcast
Data Source
AI summary
A distributor terminal includes an input unit that inputs a content that a distributor wants to distribute, a comment acquisition unit that acquires a comment given to a moving image to be distributed by a moving-image distribution server, a voice synthesis unit that generates a voice from the comment, a moving-image generation unit that generates a character content including a character or character data to perform an action according to the voice, and a moving-image synthesis unit that generates a moving image for distribution with the character content superimposed on the content.


