multimodal-art-projection/AutoMV ? reverse-engineered prompt
Reverse engineered prompt
Build me a tool that can take a full song file and automatically turn it into a coherent music video.
I want the app to analyze the music, split it into sections, figure out the mood and lyrics, then create a simple story with characters and scene ideas for each part of the song. It should generate a set of first frame images for the scenes, then turn those into video clips and stitch everything together into one final MV. If a vocal track is available, it should also support lip sync shots for some scenes.
Please keep the characters consistent across the whole video, and include a way to save the story, camera directions, character info, and final video in a results folder for each song. It’s fine if this uses online AI services and pretrained models, just make the setup clear and let me plug in my own API keys and model paths. If you need to, look up current docs online for any model or API setup.
Are you gonna build this?
make sure you review the code using coderabbit