Audio AI

Adding an AI model

To add an AI model, there's two requirements:

• Provide a definition of the model
• Provide the weights of the model

Defining the model

Create a directory in assets/audio_models with the name of your model. Inside this directory, you must have 2 files :

• model.py : This file contains the definition of your model and the processing of the data
• model.pt : This file contains the weights of your model, in the format of a PyTorch state dict.

WARNING

As of now, only PyTorch is supported, meaning all inputs to the model will be classes of PyTorch.