The TAO of Conversational AI, Recommenders, and Virtual Conferencing (GTC Spring 21 Keynote Part 8)

Опубликовано 13 апреля 2021, 18:06

In part 8 of the #GTC21 spring keynote, NVIDIA CEO Jensen Huang details how companies globally and an ecosystem of millions of developers are building their most important services on NVIDIA AI. The new NVIDIA TAO framework encapsulates the entire workflow to customize AI models. NVIDIA Fleet Command is a cloud-native platform organizations can use to securely operate and orchestrate AI across distributed fleets of computers. NVIDIA Jarvis provides state-of-the-art conversational AI that can be customized and processed anywhere. NVIDIA Merlin lets all companies benefit from the power of recommender systems. NVIDIA Maxine improves the quality of video conferencing for everyone. The NVIDIA Triton inference server handles scale-out AI services of all types -- from image and speech recognition to recommenders to language understanding.

02:44 - Watch the announcement of the NVIDIA TAO framework
03:26 - Learn what's new with NVIDIA Fleet Command
04:07 - Check out NVIDIA NGC pre-trained models and Fleet Command in action
07:05 - See the latest with NVIDIA Jarvis
12:10 - Learn about NVIDIA Merlin
12:54 - Dive into NVIDIA Maxine and watch the latest demo
16:28 - Watch the NVIDIA Triton inference server demo

Attend GTC Sessions: nvidia.com/en-us/gtc
Watch the Full Keynote: nvidia.com/en-us/gtc/keynote

Follow NVIDIA on Twitter:
twitter.com/NVIDIAGTC
twitter.com/NVIDIA
twitter.com/NVIDIAAI