Nvidia's Audio2Face is Now Open Source: Realistic 3D Avatars for Everyone
So, Nvidia just dropped a pretty big update: they're open-sourcing their Audio2Face tool. What does that mean for you? Well, if you're into 3D avatars and realistic facial animations, this is definitely something to get excited about. Basically, Audio2Face uses AI to create facial animations for 3D characters based solely on audio input. Imagine, your avatar's face mirroring the emotions and lip movements of someone speaking in real-time – pretty cool, right?
What's really interesting is that this isn't just for pre-made content. Nvidia says you can use it for live streams too. I can already picture some pretty wild applications for this, especially in the world of virtual reality and metaverse experiences.
Farm51, the folks behind Chernobylite 2, and the developers working on Alien: Rogue Incursion Evolved Edition are already playing around with it. It's a testament to the potential that these game developers are using this, isn't it?
Now, the best part. Because Nvidia is open-sourcing the software, you can tweak it to your liking. If you want to train your own models, feel free to do it, the framework is available.
Source: The Verge