Documentation
Soundbox documentation Soundbox generates audio with open-weight models on your own machine, arranges it on a timeline, and serves it over an HTTP API and to agents over MCP.
Using Soundbox Generating sound Prompt, model, seed and takes: how a generation runs, and what each parameter changes. The studio Arranging: lanes and clips, trimming, fades and crossfades, the mixer, the device rack, and the export. The library Where everything you make goes, how to find it again, and how editing works without destroying the original. Artwork Cover art for generated music, avatars for saved voices, and the image models behind them.