You can use this repo to execute NAM models with a WAV file. The main purpose of this is to convert NAM models to NeuralPi or Proteus. The generated WAV files can then be used to upload the Keith's Colab script to generate to model.
git clone https://github.com/Thomas-Sparber/NAM_to_NeuralPi
cd NAM_to_NeuralPi
git submodule update --init --recursive
mkdir Build
cd Build
cmake ..
make
This will give you the executable Build NAM_to_NeuralPi
./NAM_to_NeuralPi [path-to wav file] [NAM-model-1] [NAM-model-2] ... [NAM-model-N]
You can already use the WAV file prepared by Keith which is in the correct format to be used for NeuralPi or Proteus:
./NAM_to_NeuralPi Proteus_Capture_Utility/Proteus_Capture.wav [NAM-model-1] [NAM-model-2] ... [NAM-model-N]
The wav file will then be processed using the NAM model and the result is stored as out.wav if only one NAM model is provided or out1.wav, out2.wav, ..., outN.wav