[WIP] Music Video Generator | A New AI Orchestration System
New orchestration engine [WIP]
Here, a preview of one of its modes: "music-video".
Input: image [artist] + audio track
Output: a modest video clip.
Around 12 minutes of generation time / multi-GPU [H100s].
Hopefully, this will be all out soon. In the meantime, I'd love to read you: what is it that you'd like to see in a new online audiovisual studio?
[WIP] Music Video Generator | A New Orchestration AI System (YouTube)
Comments (6)
This is nuts! Holy shit, bravo 👏
Is it generating the video from a prompt, or the lyrics?
It's generating a video like the examples shown from a single image [of the artist] + the audio track + a directive [natural-language].
Parece estupendo pero no entiendo exactamente lo que es y como funciona...
This is great Julián! Really looking forward to seeing this. Have been thinking of developing something similar, but really I just want to make audio-visual content, so I may not go ahead with it :) I had to sign up to your Patreon, seeing you were putting this together. Nice work! I have some features in mind but don't really know how to explain it. I will be making suggestions tho, would love to see what you have already, would make it a bit easier to suggest. Anyways, looking forward to it. Thanks for this content.
Thanks, John. I'll let you know when I have a demo ready to upload.