Audio2face 3.2 Don't Work

Hello, when you start Audio 2 face 3.2 gives you this error. At the same time, the previous versions work perfectly

kit_20220207_163205.log (280.6 KB)

Hello @Wtolamich! Thanks for reporting this and attaching your log. I will have the Audio2Face team take a look and get back to you as soon as I hear back!

@WendyGram Hello, what is the team’s response? I just urgently need the function of generating blendshapes

Hi @Wtolamich. I apologize for the wait. I was informed that the development team needs your full logs found here: C:\Users\<USERNAME>\.nvidia-omniverse\logs\Kit\Audio2Face

Also if you could zip up your log file and attach them from here:

Hi @Wtolamich! I looks like from the logs you did attach that you are using an NVIDIA GeForce GTX 1660 GPU. The GTX 1660 does not have the required tensor cores needed to run Omniverse. You can take a look at our technical requirements here: Omniverse Technical Requirements

@WendyGram Thanks for the reply. But the joke is that before version 3.2 of Audio 2 face, everything worked for me and started without RTX. But I noticed that in version 3.2, the system requirements for video memory have increased (from 6GB to 8GB). I noticed this after I wrote on the forum. Well, I’ll probably have to look for a new video card

Hi @Wtolamich , thanks for the ping and sorry to hear you are having some issues.

Omniverse and Audio2Face being part of it does require RTX minimum. It is surprising it worked before. As we are using RTX and TensorRT in the GPU for the AI inference of A2F. Either way, yes you need a minimal of RTX card to use this. Hope this helps.