Nvidia gets muscle with Omniverse Audio2Face, its 3D AI tool

Nvidia has published in its blog a new post where it talks at length about the future developments that revolve around the company. As expected, artificial intelligence gets much of the attention in this post, with the Omniverse Audio2Face tool taking center stage.

This application, which is powered by AI, allows facial animations to be generated through a simple audio file. In this way, the idea is “that 3D artists can animate secondary characters efficiently, thus replacing a tedious manual process”.

According to the company, these generative models are intended to reduce the work time of 3D animators. For example, the latest version of this Omniverse Audio2Face improves quality, usability, performance and is even capable of processing numerous audio files from multiple users thanks to the REST API.

The AI, which now includes Mandarin among the available languages, looks this good in its latest released video.

This is not the first time that Nvidia has entered the field of AI; in fact, it has so much experience that it could give lessons to hundreds of companies. Back in January, we reported that they were looking into launching AI-optimized drivers to improve the performance of graphics cards.

In terms of AI, the company offers a huge suite of solutions with AI at the center of it all: free managed labs, AI-powered chatbots with Triton Inference Server, image classification and much more.