Music is an art and a universal language. It is a collection of notes of different frequencies. In this project, I have tried to develop a deep learning model that generates music on its own. The model was trained using a set of 65 tunes(in ABC notation) taken from the Nottingham Music Database. I have taken some parts of the model development code from here. The preprocessing step consisted of removing the metadata and extracting inputs and outputs from the database. I have used a series of LSTM layers along with Dropout.
