Music-Generation-using-RNN

In this project, we train a Recurrent Neural Network (RNN) to learn the structure of music files in ABC format and then we generate a music file based on what the model learns. We experiment with different parameters like dropout, changing number of neurons in the hidden layer, optimizers etc. to see their effect on model’s learning behaviour and have reported the same.