← Back to Skills Marketplace
智能音频分离工具
by
nabian1990amber-cmd
· GitHub ↗
· v1.0.0
· MIT-0
83
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install sound-split
Description
智能音频分离工具,一键将任意音频或视频分离出人声、伴奏、鼓、贝斯、钢琴等独立音轨。适用于音乐人翻唱伴奏提取、歌曲 Remix 制作、播客人声降噪、视频配乐替换、音乐教学素材准备等场景。当用户需要「分离人声和伴奏」「提取伴奏」「去除人声」「拆分音轨」「vocal split」「stem splitter」等操作时触...
Usage Guidance
This is a local web tool that matches its description, but review and take these precautions before running:
- Inspect server.py and the shipped HTML yourself (they are included) and move/copy the provided HTML into a static/ directory (server.py expects static/index.html, static/editor.html, static/landing.html) or adjust the code.
- Run the service locally (do not expose it to the public Internet). The server enables CORS for all origins and will serve files from the working directory, so keep it behind a firewall or bind to localhost only.
- Expect demucs to download a ~300–500MB model on first run (network and disk usage). Consider running in an isolated environment or VM if you are cautious about large downloads.
- The server stores uploaded files and outputs in uploads/, outputs/, trims/ and a history.json in the working directory; delete these if they contain sensitive data.
- If you plan to expose the service or use it in multi-user contexts, add authentication, tighten CORS, and validate/normalize path inputs (there is potential for serving unexpected files if endpoints are exposed).
Capability Tags
Capability Assessment
Purpose & Capability
Name/description match the implementation: server.py invokes demucs and ffmpeg to separate stems, and the shipped HTML UI supports upload, progress, preview, trimming and download. Required components (Python, FFmpeg, demucs) are expected for this purpose.
Instruction Scope
SKILL.md instructs installing Python packages and FFmpeg, running the local FastAPI server, uploading files, and using the web UI — all within scope. Notes: the skill stores uploads/outputs/history locally (uploads/, outputs/, history.json) and allows CORS origins="*" (sensible for local dev but means any page can call the API if the server is reachable). Also the SKILL.md and server expect slightly different static file locations (server.py serves static/*.html while files are provided under rules/), so users must copy/rename files before running.
Install Mechanism
No opaque installer is used. The doc recommends pip installing fastapi, uvicorn, demucs and ffmpeg-python (PyPI packages) and notes that demucs will download ~300–500MB model data into ~/.cache/demucs — this requires network access but is expected. No downloads from unknown personal servers are present in the spec.
Credentials
The skill requests no environment variables or credentials. All required external pieces (FFmpeg, Python packages) are reasonable for the stated audio processing functionality.
Persistence & Privilege
always:false and normal model invocation. The service persists tasks, outputs and history to the working directory (uploads/, outputs/, trims/, history.json) — normal for a local web tool. It does not request system-wide privileges or modify other skills.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install sound-split - After installation, invoke the skill by name or use
/sound-split - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Sound Split 1.0.0 — 首次发布!
- 提供一键分离任意音频/视频为人声、伴奏、鼓、贝斯、钢琴等独立音轨的能力
- 支持 2/4/5 轨分离模式,满足翻唱、Remix、混音和音乐教学等多场景需求
- 内含完整 Web 界面(app.html),支持文件上传、参数配置、分离进度实时反馈与音轨试听下载
- 集成 FastAPI RESTful API,支持前后端/自动化调用、历史记录管理、音轨片段裁剪与清理等操作
- 附带音频编辑器,便于在线裁剪导出指定音轨片段
- 支持主流音视频格式与高质量 MP3 输出,部署简单,功能即开即用
Metadata
Frequently Asked Questions
What is 智能音频分离工具?
智能音频分离工具,一键将任意音频或视频分离出人声、伴奏、鼓、贝斯、钢琴等独立音轨。适用于音乐人翻唱伴奏提取、歌曲 Remix 制作、播客人声降噪、视频配乐替换、音乐教学素材准备等场景。当用户需要「分离人声和伴奏」「提取伴奏」「去除人声」「拆分音轨」「vocal split」「stem splitter」等操作时触... It is an AI Agent Skill for Claude Code / OpenClaw, with 83 downloads so far.
How do I install 智能音频分离工具?
Run "/install sound-split" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 智能音频分离工具 free?
Yes, 智能音频分离工具 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 智能音频分离工具 support?
智能音频分离工具 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 智能音频分离工具?
It is built and maintained by nabian1990amber-cmd (@nabian1990amber-cmd); the current version is v1.0.0.
More Skills