An AI-powered stem splitter that isolates vocals, drums, bass, and instruments from any audio or video file locally.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install sound-split
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install sound-split using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Sound Split - AI Audio Stem Separator?
Sound Split is a professional-grade AI audio separation utility designed for musicians, producers, and content creators. It leverages advanced machine learning models to decompose a single audio or video track into distinct stems, such as vocals, accompaniment, drums, bass, and piano. As part of the Openclaw Skills ecosystem, it provides a seamless interface for local processing, ensuring data privacy while delivering high-fidelity results. This tool eliminates the need for expensive cloud services by running state-of-the-art separation models directly on your hardware.
Sound Split - AI Audio Stem Separator Use Cases
- Extracting clean instrumentals or vocals for karaoke and song covers.
- Deconstructing tracks for music production, remixing, and sampling.
- Isolating background music from video files for content editing and BGM replacement.
- Analyzing specific instrument tracks for music education, transcription, and practice.
How Sound Split - AI Audio Stem Separator Works
- Start the local FastAPI server to initialize the processing environment.
- Upload an audio or video file through the intuitive web interface.
- Select the desired number of stems (2, 4, or 5) and the AI model precision.
- The system utilizes the Demucs model to analyze and split the audio tracks in real-time.
- Preview the individual tracks in the browser and download the high-quality 320kbps MP3 outputs.
Sound Split - AI Audio Stem Separator Setup
Ensure Python 3.8+ and FFmpeg are installed on your system. Install the required dependencies using pip:
pip install fastapi uvicorn demucs ffmpeg-python
Navigate to the skill directory and launch the server:
python3 server.py
Once started, access the main interface at http://127.0.0.1:8765/app.html. The first execution will automatically download the necessary AI models (approx. 300-500MB).
Sound Split - AI Audio Stem Separator Data Schema & Taxonomy
The skill manages its workspace locally to ensure privacy and organization. Data is structured as follows:
| Path | Description |
|---|---|
outputs/{task_id}/ |
Stores generated MP3 stems like vocals.mp3 and accompaniment.mp3 |
uploads/ |
Temporary storage for user-uploaded source files |
history.json |
Persistent JSON log of the last 50 separation tasks |
trims/ |
Contains audio segments exported from the built-in trimmer |
All output files are standardized to 320kbps MP3 format for maximum compatibility.
Sound Split - AI Audio Stem Separator Advanced Features
- Integrated Audio Editor: A waveform-based visual editor to trim and export specific segments of separated stems.
- Multi-Stem Flexibility: Choose between 2-stem (Vocals/Backing), 4-stem, or 5-stem (including Piano) separation modes.
- Automated Video Extraction: Direct support for MP4, MOV, and MKV files with automatic audio stream extraction.
- Comprehensive REST API: Full programmatic control via API endpoints for developers building custom Openclaw Skills integrations.
- Local Privacy: All processing happens on the local machine with no data transmitted to external servers.
SKILL.md
Loading
Related Openclaw Skills
Text to Video on CapCut
Transform scripts and product descriptions into professional 1080p video clips using AI-powered cloud rendering.
Formula 1 Insights Skill
A comprehensive context skill providing deep technical, historical, and strategic background on the Formula 1 racing industry.
Video Editing AI App Download
A powerful AI video editing skill that automates trimming, background music insertion, and 1080p MP4 exports via a cloud-based GPU rendering pipeline.
Fudan University Knowledge Skill
A comprehensive data skill providing deep insights into Fudan University's history, academic disciplines, and global rankings.
SEO Keyword Pro
An AI-driven SEO powerhouse for discovering low-competition keywords, generating content briefs, and monitoring search engine rankings.
French Open Information Agent
An advanced research skill for retrieving comprehensive data on the French Open brand, product lines, and global market positioning.
Featured*
