Documentation
Use edytlab and build on it.
Getting Started
Install edytlab, set up your API key, and load your first audio file.
User Guide
Sessions, tracks, chat, markers, A/B compare, export, memory and skills.
Audio Tools Reference
All 113 tools the agent can call.
Developer Guide
Build from source, run tests, add tools or providers.
API Reference
Every Tauri command, bridge type and event.
About edytlab
edytlab is a local-first, open-source desktop audio editor where you chat with an AI agent to load, cut, mix, and render audio. The DSP engine runs on your machine and your audio never leaves it. The app sends the text you write to your chosen LLM provider, plus requests you start yourself: a plugin you install from a URL, an MCP server you add, the speech model you download from Hugging Face and the sign-in to OpenAI when you choose Continue with ChatGPT. See privacy.
edytlab supports six API providers plus your ChatGPT plan: Anthropic, OpenAI, Google Gemini, Groq, OpenRouter, and Ollama for a local model. You bring your own API key, stored in your OS keychain (Ollama needs none), or sign in with ChatGPT to use your Plus or Pro plan instead of a key (a preview). Switch providers any time in Settings.
Platform support
| Platform | Status | Minimum version |
|---|---|---|
| macOS (Apple Silicon or Intel) | ✅ Supported (unsigned, universal binary) | macOS 11.0 (Big Sur) |
| Windows | ✅ Supported (unsigned) | Windows 10 (with WebView2) |
| Linux | ✅ Supported (unsigned) | Ubuntu 22.04+ (.deb / AppImage) |
License and source
The source is on GitHub, released under the MIT License.