Data Player: Automatic Generation of Data Videos with Narration-Animation Interplay.

Area of Science:

• Data Visualization
• Human-Computer Interaction
• Artificial Intelligence

Background:

• Data storytelling effectively integrates visualizations and narratives, with data videos gaining prominence.
• Producing data videos requires significant manual effort and diverse professional skills, including narrative design, animation, and audio synchronization.

Purpose of the Study:

• To introduce Data Player, a novel method for automatically generating dynamic data videos with narration-animation interplay.
• To reduce the technical barriers in creating rich, narrated data videos.

Main Methods:

• Data Player extracts data into tables from visualizations.
• Large language models establish semantic connections between text and visuals.
• Animation design knowledge is encoded as constraints to recommend presets aligned with text-to-speech narration.

Main Results:

• Data Player successfully generates dynamic data videos with narration-animation interplay.
• The system demonstrates efficacy in creating high-quality data videos comparable to human-composed ones.

Conclusions:

• Data Player significantly simplifies the data video creation process.
• The automated approach lowers technical barriers, enabling wider accessibility to data storytelling through video.