code used for performance with my group Psc2
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


Psc2 combines the worlds of jazz, rock, classical music, and artificial intelligence via original compositions and arrangements.

the compositions are built around the interplay of jazz harmonies, complex rhythms and extensive improvisation; many of them are composed in the "long-form" compositional style of classical music.

the arrangements take well-known songs from pop, rock and jazz and re-build them in a manner similar to our original compositions. the arrangements include songs by michael jackson, guns n' roses, silvio rodriguez, duke ellington and others.

psc is a señor swesearcher in google brain, and is experimenting with using machine learning models (including generative models for music) as part of the live performance. most of the code used will be available in this repo.


  1. Install SuperCollider

  2. Create a virtualenv and activate it. We need Python2 (and not Python3) because pyosc is not compatible with Python3. This step is optional but recommended:

    virtualenv --system-site-packages -p python2 venv
    source venv/bin/activate
    pip install -r requirements.txt
  3. pip install pyosc.

  4. Install Tensorflow (instructions here).

  5. Install Magenta (instructions here).

  6. Clone this repo.

  7. Download Melody RNN attention_rnn model here, and update the base_models_path parameter in Psc2/modes/

  8. Open SuperCollider, open Psc2/ and run the main group (enclosed in parentheses).

  9. From the root directory, run:

    python install
  10. Start the python server:

    python Psc2/