Machine Learning tool for creative music applications
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2025-08-18 04:28:09 +02:00
synthedge initial DTW support 2025-08-18 04:28:09 +02:00
.gitignore excluded test data in gitignore 2025-07-15 09:50:15 +02:00
LICENSE Initial commit 2025-07-15 05:04:57 +02:00
main.py v0.1_release 2025-08-02 16:10:29 +02:00
osc_example.pd multi-output, loading+saving 2025-07-31 03:30:31 +02:00
README.md v0.1_release 2025-08-02 16:10:29 +02:00
requirements.txt initial DTW support 2025-08-18 04:28:09 +02:00
synthedge.ico created Icon 2025-07-31 17:07:45 +02:00

SynthEdge

Machine Learning tool for creative music applications. Inspired by WEKINATOR (https://github.com/fiebrink1/wekinator), but written in Python and on a more modern tech stack

under development

This project is under development. Use it at your own risk. Current functionality:

  • OSC Input and output
  • save and load projects
  • train and predict with the following models:
    • Classifiers:
      • MLP
      • RandomForest
      • Support Vector Machine
      • K-Nearest Neighbour
    • Regressors:
      • RandomForest Regressor
      • Linear / Polynomial Regression
      • MLP Regressor
      • Support Vector Regressor
  • model selection and OSC configurationover GUI

To do

  • dynamic time warping
  • configuration of models
  • probability
  • docs

How to use the tools

  • Docs will follow soon.
  • for now, check osc_example.pd file

How to develop

Git clone the repository Create virtual environment:

python3 -m venv .env
source .env/bin/activate

install requirements:

pip install -r requirements.txt

how to build

create virtual environment and source it(see above) build with pyinstaller main.py --onefile --noconsole --icon=synthedge.ico --add-data "synthedge.ico;."