Video generation from audio, with a twist

I have seen a lot of people mentioning and talking about video generation with stable diffusion. I really think that is the future of realistic AI videos. For now it is more of a feber dream imo. But how about the non-realistic AI video generation? I’ve tried going the opposite direction and created a tool that does lineart / doodle animated videos based on audio clips. Would love to hear what you experts has to say about it. www.storyvid.app Anything you would change or improve? All the best, Axel