The MCVAE was created for my thesis "Variational Autoencoders with Mixture Density Networks for Sequence Prediction in Algorithmic Composition". Long title.
At a top level, the thesis asks these two questions:
Does music contain a hierarchical component which is relevant when teaching a machine learning model to create music?
And, can a machine learning model learn long term structure in music, based on its own perception of data?
And the short answer two both questions is yes.
- A variational autoencoder (VAE) for sequence prediction of notes, it is composed by LSTM-layers.
- A mixture density network (MDN) comprised of a mixture model (MM) and a LSTM-network, to predict sequences of sequences of notes. Sequences of bars, to be clearer.
To test the project, follow all the steps in the notebook MCVAE full process.
To do inference, use the notebook MCVAE inference.