How to create a digital AI human Avatar

I want to create digital human AI avatars. Each with a different look. The Avitar will communicate with humans that desire to visit and explore museums. Can anyone suggest the required platform? Not clear on how to use Omniverse Avatar to accomplish my desired goal. I read::
Omniverse Avatar is a technology platform for developing interactive AI avatars. It connects NVIDIA’s core AI technologies— NVIDIA Riva for speech AI, NVIDIA Metropolis for computer vision, NVIDIA NeMo Megatron for natural language understanding, and NVIDIA Merlin™ recommendation engines.

Hello @osmondtqet6! I reached out to the development team to help answer your question!

Thanks for your assistance.

Omniverse Avatar for developing AI avatars

Just to be clear of what I need help with, and its simple to me. From the picture above, see the digital human, I want to build/design, the same. It seems that that picture was developed from a regular photo and technology added to use it.

What technology? Do I need to learn all the technologies below?

Omniverse Avatar is a technology platform for developing interactive AI avatars. It connects NVIDIA’s core AI technologies— NVIDIA Riva for speech AI, NVIDIA Metropolis for computer vision, NVIDIA NeMo Megatron for natural language understanding, and NVIDIA Merlin™ recommendation engines.

I forgot to mention a company of interest using Nvidia technology. Does Nvidia have same. Company is reallusion, worth a look.

Hi @osmondtqet6 ,

Thank you for your interesting in Omniverse Avatar.
OVA is about powering AI driven avatar. Creating or bringing in your own 3D model is current required and we work closely with many partners to help provide that.

Epic, Reallusion, Daz …etc

We provide AI based animation, Intelligence piece through Riva and NLP. We have not shipped the full solution yet and is still in development. Stay tuned for more updated announcements regarding this very soon.

Currently we do have customers if they are comfortable to take some of these technologies are use them, they can do it now. Running Riva ASR, TTS, Dialog Manager to then drive Audio2Face for an interactive experience. The process is bit more involving but doable.

For an easier solution, you would have to stay tuned for it, we will have some news on this soon.

Thanks

Thanks for your response, see attached. The answer to the 2 questions will allow us to develop our platform.

Thanks for your response.docx (267 KB)

See attached sent to Mr. Yuen. You might have the answer to the questions.

Thanks for your response.docx (267 KB)

Will answer the questions shortly.

Just additional information, if you are interested to use A2F quickly to build an interactive chatbot, here is a tutorial one of our user did. Check it out. We do plan to make the process easier, stay tuned for that.