Descript AI music and sound effects vs generating music by API

Descript's changelog for 2026-09-17 says its editor can generate music and sound effects on request. Sume's Music Router, POST /v1/music-router/generate, makes music from a prompt, and the docs describe no sound-effect route. Omit model or send sume/music-auto and Sume picks the engine, Lyria 3.5 today; the catalog also lists lyria-3-pro.

How the router takes a request

The body needs a prompt of 1 to 5000 characters. Put exclusions in the positive prompt, because a non-empty negative_prompt is unsupported. duration and duration_seconds are rejected, so steer length in the prompt, such as "a 2-minute track" or section markers like [0:00-0:30] Intro. An optional image_url is accepted. Every router model charges the fixed Music price per generation; the catalog shows each model's provider list price for reference.

Putting music under a voice

Generated tracks can go into a Timeline 1.0 render as a soundtrack with duck_db, which lowers the music under the spine. The render is $0.10 per output minute, rounded up; confirm rates in GET /v1/catalog.

Honest limits

• No effects or foley generation is documented, so use a library for those.
• Music is not length-exact; check the result duration.
• The Descript changelog does not give rights terms, so check Descript's own terms for its output.