GAN-Source-Separation Uses a generative adversarial network (GAN) to separate the vocals of a musical audio file (.wav) from the background. This was built using Keras Adapted from: https://github.com/francesclluis/source-separation-wavenet