Generative Disco: Text-to-Video Generation for Music Visualization
This addresses the time-consuming and resource-intensive task of music visualization for creative professionals, though it appears incremental as it builds on existing text-to-video methods.
The authors tackled the problem of creating music visualizations by introducing Generative Disco, a system that uses large language models and text-to-video generation to automate the process, with a study showing that design patterns like transitions and holds enabled professionals to build coherent visual narratives.
Visuals can enhance our experience of music, owing to the way they can amplify the emotions and messages conveyed within it. However, creating music visualization is a complex, time-consuming, and resource-intensive process. We introduce Generative Disco, a generative AI system that helps generate music visualizations with large language models and text-to-video generation. The system helps users visualize music in intervals by finding prompts to describe the images that intervals start and end on and interpolating between them to the beat of the music. We introduce design patterns for improving these generated videos: transitions, which express shifts in color, time, subject, or style, and holds, which help focus the video on subjects. A study with professionals showed that transitions and holds were a highly expressive framework that enabled them to build coherent visual narratives. We conclude on the generalizability of these patterns and the potential of generated video for creative professionals.