feedBack-Demucs-Server

feedBack Demucs Server is a standalone, GPU-accelerated companion service that offloads feedBack's heavy AI audio workloads — Demucs stem separation, Whisper…

  • AI
  • Media Applications
  • Tools & Utilities
feedBack Demucs Server is a standalone, GPU-accelerated companion service that offloads feedBack's heavy AI audio workloads — Demucs stem separation, Whisper lyrics alignment, and Crepe pitch extraction — to a separate CUDA-capable GPU. It runs on a dedicated desktop or GPU host while feedBack itself runs on your NAS, keeping the main app responsive. It is aimed at homelab users who want stem-splitting, lyrics alignment, and karaoke-style workflows without taxing their primary server.

Platform availability

Available on 1 of 10 platforms

  • Unraid
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

45/100 Fair

Maintained
69last commit 130 days ago
Popular
144 GitHub stars
Easy to install
351 install platform, install notes, Docker image
Light to run
No data

Based on 3 of 4 factors.

Availability: platform listings confirmed today, links checked 22 days ago

GitHub stars
4
Open issues
4
Last commit
2026-07-14
License
AGPL-3.0
Activity
stale

Checked 7 days ago - source: GitHub (got-feedback/feedback-demucs-server)

Resources & Compatibility

Requirements not published yet.

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • First launch — model download On first start the container downloads large AI model weights (Roformer, Whisper, and Crepe). Allow at least 9 GB of free disk space and give it several minutes depending on your connection speed. source
  • GPU passthrough (Unraid) To enable CUDA acceleration, install the Unraid NVIDIA Driver plugin and add --gpus all to the container's Extra Parameters. source

Fetched 47 days ago - each note links to its source

Alternatives

No alternatives collected yet.

Common questions

What should I expect on first launch of the Demucs server container?

It downloads large AI model weights for Roformer, Whisper, and Crepe. Allow at least 9 GB of free disk space and several minutes depending on connection speed.

Can I send audio to the Demucs server by URL instead of uploading a file?

Yes, the server accepts file upload or URL input. Stems are returned as lossless FLAC.

Will the server reprocess the same song if I choose a different separation model?

No, per-stem caching is keyed by both the audio and the model. The same song under two models is cached separately.

Answers sourced from github.com

Community

Collected 53 days ago - public community threads