LogInThe updated version can generate facial BlendShapes.
NVIDIA has released an update for its experimental AI-powered software for generating facial animation from audio sources Omniverse Audio2Face. The newly revealed version – Audio2Face 2021.3.2 – brings two great new features that many would find interesting.
The first feature is the new BlendShape Generation tool, which allows the user to generate a set of BlendShapes from a Neutral Head mesh. All one needs to do is to connect a character to Audio2Face using the Character Transfer process, the tool will do the rest.
The second feature that was added in this update is the Streaming Audio Player that allows the streaming of audio data from external sources such as Text-to-Speech applications via the gRPC protocol.
You can learn more about the update here. Also, don't forget to join our new Reddit page, our new Telegram channel, follow us on Instagram and Twitter, where we are sharing breakdowns, the latest news, awesome artworks, and more.