A preview of FOSS video generative AI
youtu.be
Here are some links from the description of the video:
Try it out: https://huggingface.co/spaces/Tencent... https://huggingface.co/spaces/Tencent...
It is also open source - run it locally: https://github.com/TencentARC/MotionCtrl
I wish that guy would dial his fake enthusiasm down to about 20 percent.
Haha yeah, he does that a bit too much sometimes.
Whoever it's in the video sounds like a poor AI voice over.
Sounds like Ren Hoek, You Iiiidiiiot!
You cannot have... my coveted... iiiice... creeeam... baaar!
Agreed. Narrator does sound like Ren.
I agree AND I think he really stresses the conjunctions WITH far too much inflection.
I legitimately couldn't tell if he was being serious or sarcastic the entire video.
You just have to tweak the prompt a little bit...
It’s his distinct style that he has kept since before most of us tried anything like ChatGPT etc, so personally I welcome his consistency.
So it's not a bad AI speech, it's a bad human speech? Wow.
I have enough trouble generating postage-stamp images right now. What kind of rig am I going to need for video?
First two links now return 404 errors.
All their samples are parallax effects on a still image which doesn’t really qualify as video to me.
Here is an alternative Piped link(s):
https://piped.video/2hfPVBDMB-o?si=b4X7OtoxZIpdG7mg
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
This conda thing is quite confusing since it said to not put it on the PATH, but now I can't use it via command line to set up the environment, and using the GUI I can't find the specific python version 3.10.6 they use.