Create real-time music interaction with Magenta RT
A hands-on online workshop on building real-time, AI-assisted music systems. This workshop has now taken place — the full recording is below.
Wednesday, 24th June 2026 (9am PT / 12pm ET / 5pm UK / 6pm CET)
Workshop leader: Jesse Engel (Google DeepMind, Magenta)
Watch the recording
This workshop took place on 24th June 2026. The full recording is available to anyone with a Music Hackspace account — sign in below to watch.
Watch the recording
Loading recording…
In this hands-on workshop, we explored Magenta RT — a framework for building real-time, AI-assisted music systems. We covered how to set up live generative loops, respond to musical input in real time, and integrate Magenta RT into your own creative or performance workflow.
No prior ML experience needed; a basic familiarity with music tools and Python helps when following along.
Resources
To go further with what was covered, explore the following:
- Magenta GitHub — overview of the project and codebase.
- Magenta RT intro post — original blog post introducing real-time capabilities.
- Python basics — if you're new to Python.
About the workshop leader
Jesse Engel
Jesse Engel is a lead research scientist on Magenta, a research project within Google DeepMind exploring the role of machine learning in creative applications. His research on Magenta includes creating new generative models for audio and symbolic music, including NSynth, MusicVAE, GANSynth, DDSP, MT3, SingSong, MusicLM, MusicAISandbox, and Lyria RealTime.