Installation¶

TokEye requires Python 3.11 or newer. Install the core package for the Python API and CLI, or include the app extra for the web interface:

pip install tokeye
pip install "tokeye[app]"

Heavy dependencies are split into extras so you download only what you use:

Install	What you get
`pip install tokeye`	Python API + CLI (`tokeye run`, `tokeye elmspec`). Smallest core install.
`pip install "tokeye[app]"`	+ the Gradio web app (`tokeye app`) and `.flac`/`.ogg` input.
`pip install "tokeye[ae]"`	+ torchvision, for the `ae_tf_maskrcnn` model (`tokeye alfvenspec`).
`pip install "tokeye[hdf5]"`	+ `.h5` and MATLAB v7.3 `.mat` input.
`pip install "tokeye[all]"`	All of the above.

Training is research code that runs from a clone (uv sync --group train); it is not part of the installed package.

GPU vs CPU PyTorch. A plain install pulls the default PyTorch wheel. On macOS and Windows that is already small; on Linux it is the CUDA (GPU) build (~2.5 GB). On a Linux machine without a GPU, install the CPU wheel (~200 MB) instead:

# pip: install CPU torch first, then tokeye
pip install torch --index-url https://download.pytorch.org/whl/cpu
pip install "tokeye[app]"

# uv
uv pip install "tokeye[app]" --torch-backend=cpu
uv tool install "tokeye[app]" --torch-backend=cpu
uvx --torch-backend=cpu "tokeye[app]" app