Skip to content

source-separation/tutorial

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

85 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Open-Source Tools & Data for Music Source Separation: A Pragmatic Guide for the MIR Practitioner

By Ethan Manilow, Prem Seetharaman, and Justin Salamon

This is the code repository for our ISMIR 2020 tutorial about Open Source tools for Source Separation. This repo contains the code to build the jupyter book website where the tutorial content is hosted.

Click here to visit the tutorial!

Building the book

To build the book, do the following:

pip install -e .
python -m common.data --run.cmd='download'
jb build --all book/

Running an experiment

To run a basic mask estimation experiment with a Chimera network, do the following, given the base configuration included in common/exp/conf/chimera.yml:

First, prepare the MUSDB data:

# Symlink your data directory to ./data/ and prepare it for scaper
python -m common.data --args.load=common/exp/conf/chimera.yml

Now, train, evaluate, and listen to a model:

# Train and evaluate the model
python -m common.exp.chimera --args.load=common/exp/conf/chimera.yml

Questions? Comments? Typos? Bugs? Issues?

Open a github issue here