StemStream

Extract the "stems" (e.g. drums, vocals, bass) from any song on Spotify or Youtube into separate audio files

Introduction

This is a Flask web app using the following resources:

• Demucs ML model: Demucs is an music source separation model that creates the stems
• yt-dlp: Extracts the audio from the YouTube videos before passing them into demucs.
• Spotify API & Youtube Data API for URL conversion

Screenshot

Installation

0. Install Python 3.8 or higher

1. Clone the repository: Open a terminal and clone the repository using the following command:

   git clone https://github.com/otonomee/streamstem.git
   cd streamstem
   

2. Install dependencies: Navigate to the project directory and install the required dependencies:

   pip install -r requirements.txt
   

3. Set up environment variables (Required for Spotify URL conversion):

   a. Copy the example environment file:

   cp .env.example .env
   

   b. Edit .env and add your API credentials:

   SPOTIFY_CLIENT_ID=your_spotify_client_id_here
   SPOTIFY_CLIENT_SECRET=your_spotify_client_secret_here
   GOOGLE_API_KEY=your_google_api_key_here
   

   How to get API credentials:

   • Spotify API: Register at Spotify Developer Dashboard to get Client ID and Secret
   • Google YouTube API: Create a project at Google Cloud Console and enable YouTube Data API v3

   Note: API credentials are only required if you want to use Spotify URLs. YouTube URLs work without any API keys.

4. Run the web app: Start the web app by running the following command:

   python application.py
   

5. Access the web app: Open your web browser and go to http://localhost:8001 to access the web app.

Usage

1. Enter the YouTube video URL: On the web app's homepage, enter the URL of the YouTube video you want to convert the audio from.

2. Select the number of stems you'd like:

   • 2 stems: vocals and instrumental
   • 4 stems: vocals, drums, bass, other
   • 6 stems: vocals, drums, bass, guitar, piano, other

3. Select desired output format

4. Hit Submit and wait a few minutes. Download prompt will appear when finished :)