torchsynth 0.0.1

A modular synthesizer in pytorch, GPU-optional and differentiable

torchsynth

{\tt torchsynth} is based upon traditional modular synthesis, but is GPU-enabled and is differentiable.

Development Installation

git clone https://github.com/turian/torchsynth
cd torchsynth
pip install -e .

Note that torchsynth requires PyTorch version 1.7 or greater.

Examples

We recommend that you run examples through Jupyter notebooks, and that you have jupytext installed. It's a little fiddly to install, and those instructions are the best. jupytext makes it easy to put demo notebooks into the repo as Python files. (Larger assets like ipynb files we should avoid.)

To run examples, you should also do:

pip install -e ".[dev]"

Unfortunately, Python 3.9 (e.g. OSX Big Sur) won't work, because librosa repends upon numba which isn't packaged for 3.9 yet. In which case you'll have to create a Python 3.7 conda environment. (You might also need to downgrade LLVM to 10 or 9.):

conda install -c conda-forge ipython librosa matplotlib numpy matplotlib scipy jupytext
conda install -c anaconda ipykernel
python -m ipykernel install --user --name=envname

and change the kernel to envname.

Tests

Unit testing is performed using pytest.

pytest and other project development dependencies can be installed as follows:

pip install -e ".[dev]"

To run tests, run pytest from the project root:

pytest

To run tests with a coverage report:

pytest --cov=./src