Skip to main content
Use the client from request setup. These examples assume api, userAddress, and the returned songId are available. api includes the session token and selected account headers.

​
Persist a track mixer

volume is linear gain: 1 is unity, 0 is silence. pan ranges from -1 to 1. Apply those settings in your player. This route does not bake a mix into the song URL.

​
Save a snapshot

yourTracks and yourClips are your editor’s serializable state. Omit timelineSnapshot to save the API’s current stems and active-take information. Read /api/editing/history/{songId} for newest-first commits. A snapshot records state; it does not execute undo or restore your editor. To revisit a snapshot, your UI loads its supported fields and renders the arrangement. Use take activation to persist a saved audio selection.

​
Keys, drums, samples, live code and effects

These controls belong to the browser workstation. There are no public endpoints for playing a key, submitting a Strudel pattern, sequencing drums or applying its effect rack. Your product can build those parts with its own audio tools. Render them to a valid WAV, upload the source, then pass the returned url as srcAudioUrl to an edit. Store your editor state in your own system or a timeline snapshot. For a complete custom bounce, upload its WAV and PATCH the editable song’s url and duration. That metadata update does not create a candidate take or synchronize every stem; keep your arrangement records consistent and save a snapshot. Use the candidate merge flow when merging eligible API takes.