I think Mike Boyd did a video where he learnt how to create a deep fake and it took him 100 hours. He used deepfakelab and 3 rendering PCs just to make some short clips. It would take a lot of rendering to do an entire show.
Here is an alternative Piped link(s):
https://piped.video/Zmutd9618Kk?si=xTgB7An2gDqyQ3SX
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source; check me out at GitHub.
Just do it. It’s not hard, it just takes time to learn all the pieces and how they fit together.
im entirely deepfake now it’s so cheap
At least you’re a deepfake. They ran out of of CPU cycles in their cloud account, so I’m just a shallowfake. :/
It’s porn isn’t it.
Asking for a friend
It’s totally not!
Charlie Chaplin does Dallas
Alright alright alright
Now it makes me want to swap the actor who played Jafar in Aladdin (2019) by Jason Mantzoukas and see how better the movie becomes.
Which NOT PORN show were you watching? Which actor did you want to replace?
AI art is very easy to learn, but has a lot of intricacies and extra functions that can take a while to learn and master. The only free and open source one, Stable Diffusion, has a lot of resources online. Here’s a beginner’s guide to get you started.
I recommend running locally if you have the hardware for it. You generally need at least 8GB of VRAM for reasonably fast generation, though I’ve heard of people being fine with as little as 4GB. Check out Civitai for additional resources.
Bro OP’s not talking about genAI. OP’s talking deepfake. Big difference. You should be pointing them to DeepFaceLab, not SD.
