Contemporary music genre rhythm generation with machine learning

Abstract

In this article, we present research on customizing a variational autoencoder (VAE) neural network to learn models and play with musical rhythms encoded within a latent space. The system uses a data structure that is capable of encoding rhythms in simple and compound meter and can learn models from little training data. To facilitate the exploration of models, we implemented a visualizer that relies on the dynamic nature of the pulsing rhythmic patterns. To test our system in real-life musical practice, we collected small-scale datasets of contemporary music genre rhythms and trained models with them. We found that the non-linearities of the learned latent spaces coupled with tactile interfaces to interact with the models were very expressive and lead to unexpected places in composition and live performance musical settings. A music album was recorded and it was premiered at a major music festival using the VAE latent space on stage.