gan#samples

a nn to generate new audio samples from scratch (currently 16khz)

start

1. add your training data (.wav files) to training_data/raw/
2. run process_raw_data.py
3. you should now have 16hz versions of your training data named and numbered inside training_data/processed/