How to Install Audio2Face in a Docker Container When Installer is Unavailable

Omniverse Apps Audio2Face
, ,
1

I’m trying to set up a Docker container to run Audio2Face. I have successfully installed the required prerequisites, such as the NVIDIA Toolkit and NVIDIA Docker containers, and Cuda. I can run GPU containers fine. However, I’m unable to find the Audio2Face installer for Linux to actually run inside my Docker container.

I also found a video here in your official Nvidia playlist

It suggests to use docker pull nvcr.io/eevalgoeixww/staging/audio2face_pipeline:0.5.0, but also infers that you need access to ACE. I’m currently not using ace and don’t have access to it. I’m just using Audio2face with my own 3D models. Is this suggesting that audio2face cannot be run in a container, unless you have Nvidia Avatar Creation Engine access?

I’m stuck on how to proceed with installing Audio2Face itself as I don’t have the installer. Is there a way to obtain the Linux installer for Audio2Face, or is there a different approach to installing it in a Docker container? Do I have to have audio2face installed in the wsl 2 host machine, for it to run in container?

Any guidance would be appreciated.

Thank you!
Ej

2

Unfortunately, Docker is only available to ACE users.

3

So how am I supposed to run it on the cloud?

4

Audio2Face can run on the cloud but non-ACE users don’t have access to that docker image. That said, you could create the docker image yourself, based on your needs.

Please note that the free Audio2Face (non-ACE) can be only used by 2 people and you probably want to spawn multiple instance of A2F which violates the licensing. We will be able to assist you better if we understand your use case better.