PyTorch
demucs
audio-separation
music-source-separation
music-source-separation-model
source-separation
audio-source-separation
stem-separation
music-separation
vocal-separation
vocal-isolation
vocal-removal
vocal-extraction
instrumental-separation
instrumental-extraction
instrumental-isolation
karaoke
karaoke-maker
acapella
acapella-extraction
vocals
voice-isolation
drum-separation
drums-isolation
bass-separation
bass-isolation
guitar-separation
piano-separation
instrument-separation
multi-stem-separation
two-stem-separation
four-stem-separation
six-stem-separation
htdemucs
htdemucs-ft
htdemucs-6s
mdx-extra
hybrid-transformer
transformer
deep-learning
machine-learning
artificial-intelligence
generative-ai
neural-networks
audio-ai
music-ai
music-processing
audio-processing
audio-analysis
digital-signal-processing
dsp
audio-enhancement
audio-extraction
audio-manipulation
audio-conversion
audio-transcription
audio-tools
music-tools
music-production
music-editing
music-remix
remix-tools
audio-effects
audio-export
audio-downloader
mp3
wav
flac
mp4
mkv
mov
ffmpeg
media-processing
video-audio-extraction
video-to-audio
video-processing
multimedia
audio-file-converter
bitrate-control
volume-boost
batch-processing
zip-download
google-colab
colab-notebook
jupyter-notebook
google-colab-gpu
nvidia-t4
gpu-acceleration
cuda
cloud-gpu
python
python-project
model-inference
inference
open-source-ai
ai-tools
audio-separation-tool
vocal-remover-tool
music-separation-tool
audio-separation-tutorial
tutorial
beginner-friendly
no-code
low-code
interactive-ui
ipywidgets
ai-workflow
machine-learning-project
ai-project

🎢 Demucs in Google Colab: Free Vocal & Stem Separation

This repository contains an easy-to-use Google Colab notebook for running Demucs. This AI tool allows you to isolate vocals, drums, bass, and other instruments from any audio or video file for free using a Google Colab T4 GPU.

πŸŽ₯ Watch the Tutorial: Setup Demucs in Google Colab

πŸš€ Run in Colab: Open Google Colab Notebook

Open In Colab

Get Pro


✨ Features Supported in this Notebook

This notebook makes it incredibly simple to use Demucs through an interactive UI:

  • Media Support: Upload audio (.mp3, .wav, .flac, etc.) or video files (.mp4, .mkv, .mov, etc.) directly. The tool automatically extracts the audio track from videos.
  • Multiple Models: Choose between htdemucs, htdemucs_ft, htdemucs_6s, and mdx_extra.
  • Separation Modes: Extract 2 stems (e.g., Vocals + Instrumental), 4 stems, or up to 6 stems (Vocals, Drums, Bass, Other, Guitar, Piano).
  • Audio Output Settings: Export as mp3, wav, or flac with adjustable bitrates.
  • Volume Boost: A built-in slider (0-20 dB) allows you to automatically boost quiet background/instrumental tracks.
  • Automatic Zipping: Packages all your separated stems into a .zip file and downloads it automatically.

πŸ› οΈ How to Use

  1. Click the "Open in Colab" badge above.
  2. Go to Runtime > Change runtime type and ensure a T4 GPU is selected.
  3. Step 1: Click the play button on Cell 1 to install Demucs, set up FFmpeg, and check your GPU status.
  4. Step 2: Scroll to Cell 2 and choose your settings (Model, Separation Mode, Output Format, etc.).
  5. Click the play button on Cell 2. A button will appear asking you to upload your files.
  6. The AI will extract the audio, separate the stems, and display audio players so you can preview the results immediately! Finally, it will automatically download your tracks in a ZIP file.

🀝 Credits

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support