Hello Farcasters! Has anyone seen functional “text to video” artificial intelligence yet? Not tools for work, but more “Midjourney for video”? I’m assuming Midjourney and OpenAI will have this eventually, just eager for something soon!

perl.xyz is your onchain social arcade, powered by $DEGEN. Join the discussion at /perl

Building something new. 🟪 Host: /venturecapital /adventures /theitinerant

Would love to learn more! Thanks @benersing

Tinkering. Co-Founder @ Livepeer. https://www.yondonfu.com/

Not quite pure text2video, but https://github.com/nateraw/stable-diffusion-videos is a cool way to generate music videos that "morph" between the images from two prompts.

Helping pioneer startup fundraising using NFTs. Follow me here, on Twitter and/or on YouTube for advice & resources.

Yeah I do believe I saw that in a recent video... RunwayML seems to be the closest to this, though I think they still use stock video? https://www.youtube.com/watch?v=oKzAuJRPaIY&t=1s

Thanks @nebula! Would LOVE to learn more @fabian!

yeah also been dying for this. deforum for SD can do this but i haven’t figured it out yet

https://www.synthesia.io/ for text2speech with their range of avatars

Text->Video at the same level as Midjourney is not there yet. Some papers with interesting results have started to come out though. Like this one: https://github.com/showlab/Tune-A-Video