Setting up Audio2Face without RTX

Hello there,

I need Audio2Face’s Audio to Lip Sync feature for a personal project, but I only have a GTX 1660 in my machine. I installed Omniverse as a test and added Audio2Face, but, as expected by having read the minimum requirements, Audio2Face does not load properly with my GPU.

I don’t want to purchase an RTX specifically for this, so I’m wondering how I can alternatively use Audio2Face. I’m new to the whole Nvidia ecosystem so my question may seem silly: Is there any way to somehow access the prerequisite hardware via VMs or similar so that I can use Audio2Face?

Thank you for any information on this!

I also have a sane question. I donot have a RTX PC, and could not buy. Wonder if i can use cost-effective cloud environment where I can run A2F face. I tried with Google cloud. But I am not sure exactly which GPU I need to select. Also wonder if there are other cloud options. Best,

Hi @Benutzer5373 and @user133956! In order to render scenes in Omniverse, it requires the Ray Tracing Tensor cores found only on our RTX GPUs. Here is a link to our technical requirements: Technical Requirements — Omniverse Create documentation

To answer your question about Cloud options, @user133956 , we just announced Omniverse Nucleus Cloud enables “one-click-to-collaborate” simple sharing of large Omniverse 3D scenes, meaning artists can collaborate from across the room or the globe without transferring massive datasets. Changes made by the artist are reflected back to the client — like working on a cloud-shared document — but for a 3D scene.

Hi @WendyGram! So just so I understand: In the future, I could use Omniverse Nucleus Cloud to be able to run Audio2Face via the cloud? Ideally, I would send audio files automatically via a Python script and receive the corresponding face animations as output.

It also seems like I would need to register for early access to Omniverse Nucleus Cloud. I would use Audio2Face for a private project and so presumably I would have to wait until the full release of the cloud, right?

1 Like

Ideally yes, you should be able to send audio files and receive facial animations through the Cloud Service. No need to wait! I would love if you would participate in the early access program and try this out! Let us know what your experience is so that we can make sure this functionality works when we do the full release!


Hello, I’ve signed up for the program. Can you please tell me how to upload my audio file?

Hello, I would be interested in uploading audiofiles to the nvidia-cloud to process/generate face-animations with them, and download/import them into Blender. How could I do that please?

Hello @chrissie1! You can sign up for NVIDIA Omniverse Cloud here: Experience Omniverse in the Cloud | NVIDIA

If you have an RTX compatible card, you can download Omniverse for free from here: Omniverse Platform for 3D Design Collaboration and Simulation | NVIDIA and start using Audio2Face now!

1 Like