09.11.2021 at 08:00 pm
NVIDIA has announced a new platform for creating avatars using artificial intelligence called Omniverse Avatar.
The platform combines a number of separate technologies, including speech recognition, artificial speech, face tracking, 3D animation, computer vision, natural language understanding, recommendation engines and simulation technologies, which the company says can be used to build a portfolio of virtual agents.
The Omniverse Avatar opens the door to creating AI assistants that can be easily customized for almost any industry.
This can help with billions of daily customer service interactions, restaurant orders, banking transactions, appointments, personal reservations, and more, leading to increased business opportunities and improved customer satisfaction.
At the annual GTC conference, the company’s CEO shared several examples via Omniverse Avatar: Project Tokkio for customer support, DRIVE Concierge AI for always-intelligent services in vehicles, and Project Maxine for video conferencing.
In Project Tokkio’s first demo, he highlighted an avatar of customer service in a restaurant kiosk, able to see, speak and understand two customers as they order veggie burgers, fries, and drinks.
The character uses face-tracking technology to maintain eye contact with customers and respond to their facial expressions. This is useful in smart retail and customer service, the company said of the technology.
In Project Tokkio’s second demo, an animated version of the company’s CEO answered questions on topics including climate change and protein production.
In a third demo, someone used a realistic animated avatar of themselves during a conference call. The caller was dressed casually in a crowded cafe. But his default avatar was elegantly dressed and speaking without any background noise.
This example is based on Project Maxine for video conferencing from Nvidia. Which aims to improve common issues in video conferencing (such as low-quality broadcasting and maintaining eye contact) with the help of machine learning fixes.
Want to fill virtual and physical worlds with avatars:
In a demonstration of the DRIVE Concierge AI platform. The digital assistant located on the central dashboard display helps the driver determine the best driving position to reach his destination on time. It then follows his request to set a reminder once the vehicle’s range falls below 100 miles.
In the presentation, the company noted that the avatar has a response time of two seconds, slower than a human. It is bound to cause frustration if customers are in a hurry.
The Omniverse Avatar announcement comes as part of the company’s vision, called NVIDIA Omniverse. It is a virtual world collaboration and simulation platform for 3D workflows. It is currently in open beta with more than 70,000 users.
Like Metaverse, Omniverse is primarily about shared virtual worlds that allow remote collaboration. But compared to the vision put forward by Meta, the owner of Facebook. NVIDIA is less interested in bringing your office meetings to virtual reality. And more interested in copying industrial environments with virtual ones and creating avatars that interact with people in the physical world.
Written by:
CEO of OTTO Magazine