A text-based step sequencer
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.



Build Status

Jez is a text-based step sequencer for generative music.

Jez implements a domain specific language for expressing step sequences and provides a command-line tool for their playback over Open Sound Control and MIDI.

Nested step sequences are supported, for creating more rhythmically complex patterns, and a range of built-in keywords and track effects are provided for adding further variation.


  • Deterministic.
  • Support for MIDI & Open Sound Control (over UDP and WebSockets).
  • Range of built-in keywords & track effects for manipulating sequences.
  • Support for user defined keywords.
  • Light on system dependencies.


  • C API for FFI usage (primarily for Web Assembly).
  • MIDI clock support & other forms of synchronisation.
  • Better documentation and examples.