Failed to build the TensorRT engine in Audio2Face

AI & Data Science Digital Humans (closed) Audio2Face (closed)
1

Hello to everyone.

So I was using the Audio2Face program, but my system couldn’t generate the TensorRTEngine. The face model loads flawlessly and everything is fine, but there is no movement in the face model while playing the audio recording.

System Specifications:
RTX 3050 with 16GB RAM
Audio2Face 2022.2.0

I tried several times to reinstall it, but nothing worked.

I am also attaching the log file here
kit_20230321_012109.log (487.1 KB)

2

Looks like a duplicate of this thread: Failed to build TensorRT engine - Apps / Audio2Face - NVIDIA Developer Forums

3

Hello @akramkhan! Looking through your logs, it looks like your computer is not picking up your GPU. It says you only have Intel(R) UHD Graphics. Could you check to make sure your RTX 3050 card is connected to your computer and try again?

4

Yes, @WendyGram the RTX 3050 graphics card is attached, as I have verified numerous times, but I am unsure of why the Audio2face software is unable to recognise it. I’ve done everything I can, but it seems to be ineffective.

5

Have you installed graphics driver version 527.37?

6

Yes @Ehsan.HM the said driver is installed on the device. I even deleted and reinstalled it again but it seems to be of no use

7

Now that you have the driver installed, could you please share your latest log file again?