Melody Generation Using Neural Networks

The goal of this project is to train an artificial neural network (ANN) to learn how melodies are formed by making transitions between scale tones within a musical key. A user of this system supplies a collection of midi files containing songs of their choosing and another midi file (or manually inputted melody) which acts as the test data for the network. The network learns how to transition between scale degrees using the Library and then generates a new melody based on the manually inputted melody which is inspired by the testing data. The user is able to save trained models as well as output files and view/listen to them