search

NVIDIA-AI-Blueprints / digital-human

NVIDIA AI Blueprint for digital human for customer service.
https://build.nvidia.com/nvidia/digital-humans-for-customer-service
Other
130 stars 71 forks source link

Where's James? #10

Open pythonllm opened 1 month ago

pythonllm commented 1 month ago

Good afternoon. I ran the project on my server, it works I played around and customized other scenes except Ben Can you please tell me if I can get James somewhere? You wrote so much about him that I would like to run him on my server Should I create a 3d head model myself or use one of the ones you offer in audio2face?

Where can I find James?

siruslan commented 1 month ago

Exactly! It’s very confusing at this page https://build.nvidia.com/nvidia/digital-humans-for-customer-service/blueprintcard

“This experience showcases James, our interactive digital human that has the knowledge of NVIDIA’s products by having direct access to our product knowledge base. James and the RAG-powered backend application use a collection of NVIDIA NIM inference microservices, NVIDIA ACE technologies, and ElevenLabs text-to-speech to provide natural and immersive responses. Using James as an inspiration, users would be able to download and customize the Digital Human for Customer Service blueprint for their industry, with document ingestion from RAG and customizing the avatar look and voice for their application.“

It speaks a lot about James, but there is no James.

dekishan07 commented 3 weeks ago

Hi @pythonllm. I assume you manage to deploy this on your on-prem GPU server. Can advise how did you achieve this? Thanks

pythonllm commented 3 weeks ago

Hello) @dekishan07 , I deployed it not on my local machine, but on the Google server according to this guide:

https://docs.nvidia.com/ace/latest/workflows/tokkio/text/Tokkio_GCP_CSP_Setup_Guide_automated.html

or what You mean?

dekishan07 commented 2 weeks ago

Hi @pythonllm. Thanks for the update. I am trying to install it on my on-prem GPU server. I thought you were referring to on-prem GPU server when you mentioned "my server".