Skills
.Work. You
Rest

demucs-separation

Community

GPU audio stem separation in real time

Software Engineering#libtorch#demucs#gpu-inference#htdemucs#audio-stem-separation#real-time-preview
AuthorSpiralCloudOmega
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Demucs/HTDemucs-based AI stem separation enables producers, audio engineers, and developers to extract individual instrument tracks from a mixed audio file for remixing, mastering, and analysis.

Core Features & Use Cases

  • 4-stem and 6-stem separation for drums, bass, vocals, and other instruments, with guitar and piano added in the 6-stem model
  • GPU-accelerated inference via LibTorch for native performance without Python runtime
  • Real-time preview mode for interactive auditioning before committing to full separation
  • Overlap-add segment processing for long tracks to avoid boundary artifacts
  • Python script example for batch processing and saving individual stems

Quick Start

Run a separation on a mixed audio file to produce 4 or 6 stem WAVs and preview the results in real time

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: demucs-separation
Download link: https://github.com/SpiralCloudOmega/DevTeam6/archive/main.zip#demucs-separation

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.