superdirtpy

superdirtpy is a client library for Super Dirt written in python.

features

• timing control
• no framework or DSL
• can be used for algorithmic composition, live coding, integration with python libraries such as numpy and scipy and developing tools for your own use.

prerequisite

Install Tidal Cycles. Especially Super Dirt.
https://tidalcycles.org/

examples

import logging

import numpy as np

import superdirtpy as sd

rng = np.random.default_rng()
client = sd.SuperDirtClient()
s = "mydevice"
midichan = 0


def main():
    tctx = sd.TemporalContext()
    scale = sd.Scale(sd.PitchClass.C, sd.Scales.messiaen3)
    for _ in range(16):
        n = rng.integers(1, 5, endpoint=True)
        chord = rng.choice(10, n, replace=False).tolist()
        chord = scale.bind(degrees=chord)
        params = {
            "s": s,
            "midichan": midichan,
            "n": [chord],
            "amp": rng.uniform(0.4, 0.9),
            "delta": rng.uniform(0.8, 2.0),
        }
        sd.Pattern(client=client, params=params).play(tctx)


if __name__ == "__main__":
    try:
        logging.basicConfig(level=logging.DEBUG)
        main()
    except KeyboardInterrupt:
        pass

See more examples.

parameters

This library internally handles only s, sound, n, and delta, but you can also use parameters such as amp and room. (source)