Run the following command to run Audio2Face-3D:

docker run --rm --network=host -it --gpus all nvcr.io/nvidia/ace/audio2face:1.0.11

Generating TRT models:

./service/generate_trt_model.py built-in claire_v1.3
./service/generate_a2e_trt_model.py

In service/a2f_config.yaml, make sure to set the IP (default 0.0.0.0) under grpc_output to your system’s IP.

nano service/a2f_config.yaml

Launch service:

./service/launch_service.py service/a2f_config.yaml

The following log line appears when the service is ready:

[2024-04-23 12:44:33.066] [  global  ] [info] Running...

Note

More details on Audio2Face-3D deployment can be found in Container Deployment.