Baseline code and model, as well as MIDI dataset is taken from https://github.com/corynguyen19/midi-lstm-gan, which tackles a similar problem to ours. We expand on this by testing the use and efficiency of GRUs as plausible substitutes to GANs employing LSTM-based RNNs.
bjmedina / gr-gan Goto Github PK
View Code? Open in Web Editor NEWUsing GANs to create music in response to a query