deep griffinlim iteration - michele-perrone/SpectrogramPlayer Wiki

Deep GriffinLim Iteration:

Here we describe the steps to train the DGL model. Minor changes were made, w.r.t. the original Repo, to accomodate the use of multiple dataset and update legacy libraries.

Folder structure

  • model: contains the DGL model
  • where to preprocess the dataset to be ready for the training. To process the files enter: python TRAIN/TEST --num_snr YOUR_CHOICE. We tested our model with num_snr=3.
  • here the individual wav results are loaded and saved in a single numpy array. This way, we can utilize them directly in our main Jupiter Notebook.
  • used by other modules, subclass of the Pytorch Dataset class.
  • where paths are set as well as training/testing conditions and other general parameters.
  • here the training/testing conditions are set. To train or test use: python --train/test
  • this module writes the logs of the training, as well as the result wav files.
  • contains the code for training and testing.
  • various methods used by the modules.
⚠️ ** Fallback** ⚠️