Nvidia’s latest Omniverse tool creates AI avatars that can see and speak

Nvidia's Omniverse is full of interesting tools and applications for creators, designers, researchers and engineers. The latest addition to its toolset is Omniverse Avatar, a technology platform to create interactive AI avatars that see, speak, and interpret spoken intent, all rendered with ray-traced 3D graphics.

Omniverse Avatar opens the door to the creation of AI assistants that are easily customizable for virtually any industry. It could handle restaurant orders, banking transactions, making personal appointments and reservations, and more.

According to Jensen Huang, Nvidia's CEO, Omniverse Avatar combines the company's experience in foundational graphics, simulation and AI technologies so users can create collaborative robots and virtual assistants for a business or just to have some fun. During the tool's announcement, Jensen mentioned and showcased some examples of virtual assistants like Project Tokkio for customer support, Nvidia Drive Concierge for always-on and intelligent services in vehicles, and Project Maxine for video conferencing.

To create an avatar, Omniverse's latest tool uses speech AI capabilities of Nvidia Riva, computer vision and the natural language understanding Megatron 530B framework. Moreover, it also employs Nvidia's Merlin recommendation engine and Video2Face and Audio2Face facial animation technologies. Anyone interested in Nvidia Omniverse can download it HERE.

Discuss on our Facebook page, HERE.

KitGuru says: With Nvidia Omniverse is combining a lot of Nvidia's technologies in some very interesting ways. What do you all think of the new Avatar showcase? 

Become a Patron!