Spleeter

Spleeter is the Deezer audio source separation library with pretrained models written in Python and uses Tensorflow. The instrument is able to divide a music track into separate components (vocal, drums, bass and others specifical sounds). After splitting into several audio tracks, each of them can be used for their own purposes (remove vocals, cut guitar rhythm and more). The resulting audio tracks can be imported into any audio editor, such as Audacity.It makes it easy to train source separation model (assuming you have a dataset of isolated sources), and provides already trained state of the art model for performing various flavour of separation :

Vocals (singing voice) / accompaniment separation (2 stems) Vocals / drums / bass / other separation (4 stems) Vocals / drums / bass / piano / other separation (5 stems)

2 stems and 4 stems models have state of the art performances on the musdb dataset. Spleeter is also very fast as it can perform separation of audio files to 4 stems 100x faster than real-time when run on a GPU.

We designed Spleeter so you can use it straight from command line as well as directly in your own development pipeline as a Python library. It can be installed with Conda, with pip or be used with Docker. Quick start

Want to try it out ? Just clone the repository and install a Conda environment to start separating audio file as follows:

git clone https://github.com/Deezer/spleeter conda env create -f spleeter/conda/spleeter-cpu.yaml conda activate spleeter-cpu spleeter separate -i spleeter/audio_example.mp3 -p spleeter:2stems -o output

You should get two separated audio files (vocals.wav and accompaniment.wav) in the output/audio_example folder.

For a more detailed documentation, please check the repository wiki.

There is also an unofficial website that allows you to use Spleeter online: https://melody.ml