Further reading

• MusicVAE: Creating a palette for musical scores with machine learning: Magenta's team blog post on MusicVAE, explaining in more detail what we've seen in this chapter (magenta.tensorflow.org/music-vae)
• A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music: Magenta's team paper on MusicVAE, a very approachable and interesting read (arxiv.org/abs/1803.05428)
• GrooVAE: Generating and Controlling Expressive Drum Performances: Magenta's team blog post on GrooveVAE, explaining in more detail what we've seen in this chapter (magenta.tensorflow.org/groovae)
• Learning to Groove with Inverse Sequence Transformations: Magenta's team paper on GrooVAE, very approachable and interesting read (arxiv.org/abs/1905.06118)
• Groove MIDI Dataset: The dataset used for the GrooVAE training, composed of 13.6 hours of aligned MIDI and...