Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

Music Demixing Banner

Music Demixing Challenge - Starter Kit


This repository is the Music Demixing Challenge Submission template and Starter kit! Clone the repository to compete now!

This repository contains:

  • Documentation on how to submit your models to the leaderboard
  • The procedure for best practices and information on how we evaluate your agent, etc.
  • Starter code for you to get started!

NOTE: If you are resource-constrained or would not like to setup everything in your system, you can make your submission from inside Google Colab too. Check out the beta version of the Notebook.

Table of Contents

  1. Competition Procedure
  2. How to access and use dataset
  3. How to start participating
  4. How do I specify my software runtime / dependencies?
  5. What should my code structure be like ?
  6. How to make submission
  7. Other concepts
  8. Important links

Competition Procedure

The Music Demixing (MDX) Challenge is an opportunity for researchers and machine learning enthusiasts to test their skills by creating a system able to perform audio source separation.

In this challenge, you will train your models locally and then upload them to AIcrowd (via git) to be evaluated.

The following is a high level description of how this process works

  1. Sign up to join the competition on the AIcrowd website.
  2. Clone this repo and start developing your solution.
  3. Train your models for audio seperation and write prediction code in
  4. Submit your trained models to AIcrowd Gitlab for evaluation (full instructions below). The automated evaluation setup will evaluate the submissions against the test dataset to compute and report the metrics on the leaderboard of the competition.

How to access and use the dataset

You are allowed to train your system either exclusively on the training part of MUSDB18-HQ dataset or you can use your choice of data. According to the dataset used, you will be eligible for different leaderboards.

👉 Download MUSDB18-HQ dataset

In case you are using external dataset, please mention it in your aicrowd.json.

  "external_dataset_used": true

The MUSDB18 dataset contains 150 songs (100 songs in train and 50 songs in test) together with their seperations in the following manner:

├── train
│   ├── A Classic Education - NightOwl
│   │   ├── bass.wav
│   │   ├── drums.wav
│   │   ├── mixture.wav
│   │   ├── other.wav
│   │   └── vocals.wav
│   └── ANiMAL - Clinic A
│       ├── bass.wav
│       ├── drums.wav
│       ├── mixture.wav
│       ├── other.wav
│       └── vocals.wav

Here the mixture.wav file is the original music on which you need to do audio source seperation.
While bass.wav, drums.wav, other.wav and vocals.wav contain files for your training purposes.
Please note again: To be eligible for Leaderboard A, you are only allowed to train on the songs in train.

How to start participating


  1. Add your SSH key to AIcrowd GitLab

You can add your SSH Keys to your GitLab account by going to your profile settings here. If you do not have SSH Keys, you will first need to generate one.

  1. Clone the repository

    git clone
  2. Install competition specific dependencies!

    cd music-demixing-challenge-starter-kit
    pip3 install -r requirements.txt
  3. Try out random prediction codebase present in

How do I specify my software runtime / dependencies ?

We accept submissions with custom runtime, so you don't need to worry about which libraries or framework to pick from.

The configuration files typically include requirements.txt (pypi packages), environment.yml (conda environment), apt.txt (apt packages) or even your own Dockerfile.

You can check detailed information about the same in the 👉 file.

What should my code structure be like ?

Please follow the example structure as it is in the starter kit for the code structure. The different files and directories have following meaning:

├── aicrowd.json           # Submission meta information - like your username
├── apt.txt                # Packages to be installed inside docker image
├── data                   # Your local dataset copy - you don't need to upload it (read
├── requirements.txt       # Python packages to be installed
├──                # IMPORTANT: Your testing/prediction code, must be derived from MusicDemixingPredictor (example in
└── utility                # The utility scripts to provide smoother experience to you.

Finally, you must specify an AIcrowd submission JSON in aicrowd.json to be scored!

The aicrowd.json of each submission should contain the following content:

  "challenge_id": "evaluations-api-music-demixing",
  "authors": ["your-aicrowd-username"],
  "description": "(optional) description about your awesome agent",
  "external_dataset_used": false

This JSON is used to map your submission to the challenge - so please remember to use the correct challenge_id as specified above.

How to make submission


Best of Luck 🎉 🎉

Other Concepts

Time constraints

You need to make sure that your model can do audio seperation for each song within 4 minutes, otherwise the submission will be marked as failed.

Local Run



🙏 You can share your solutions or any other baselines by contributing directly to this repository by opening merge request.

  • Add your implemntation as test_<approach-name>.py
  • Test it out using python test_<approach-name>.py
  • Add any documentation for your approach at top of your file.
  • Import it in
  • Create merge request! 🎉🎉🎉


📎 Important links

💪  Challenge Page:

🗣️  Discussion Forum:

🏆  Leaderboard: