AI music for one section of a video: no duration field
Sume's Music 1.0 rejects duration and duration_seconds. Ask for the length in the prompt, then loop and fade the track to fit one section on a Timeline.

To get background music for one section of a video from Sume, describe the length in the prompt and fit the result on a Timeline. The API has no duration field: Music 1.0 rejects duration and duration_seconds, so for a 12-second section you ask for a short track in the prompt, then set the Timeline length and use its soundtrack fields (loop, fade out) to fit the music to the section.
Descript's changelog of 2026-09-17 says its AI drops generated music onto your timeline at the length you ask for. On Sume that fitting step is explicit, using Music 1.0 and Timeline 1.0.
How do I ask for a length?
The Music docs say the model is steered by the prompt: say "a 2-minute track", or use section markers like [0:00-0:30] Intro: .... Write the length and the mood in the same sentence, and add exclusions such as "no vocals" as positive text, since negative prompts are not supported. The docs do not promise the output hits the stated length to the second, so treat it as a target and let the Timeline's audio.duration_seconds set the final length.
Which route should I call?
Music 1.0 is retiring gradually. Its routes still work, but every request now resolves through the Music Router, and the docs say new integrations should call POST /v1/music-router/generate. Pricing for the music route is $0.125 per audio, the same however long the prompt is.
How do I fit the track to the section?
Timeline 1.0 has an optional soundtrack bed. Its fields decide how the music behaves under your section.
| Field | Effect |
|---|---|
url | The Sume-hosted music file |
gain_db | Level of the bed |
loop | Repeat the track if it is shorter than the program |
fade_out_seconds | Fade at the end, up to 10 seconds |
duck_db | Lower the bed under speech, 0 to 20 dB; needs a real spine, not silence |
What if the section is very short?
Set audio.duration_seconds on the Timeline to the length you need and a fade_out_seconds that suits it. If you want a track for a whole video rather than one section, see add background music to a video.
Sources
Related posts
More in Use cases
- AI object remover from a photo: mask_url on GPT Image 2.5
To remove an object from a photo through the Sume API, send the photo and a public mask_url to POST /v1/images with a GPT Image 2.5 model, then prompt the fill.
- AI product color changer: colorways through an image edit API
To recolor a product photo, send it as an input_references image with a prompt naming the new color. Sume's edit calls take up to 10 images per request.
- Amazon AI-generated people tag: contains-synthetic-performer
Amazon's seller post says photorealistic AI people in listings, A+ and shoppable videos need the XMP keyword contains-synthetic-performer. Tag the final file.
- AI Act Article 50(2): 2 December 2026 grace period, in plain terms
Commission FAQ: providers of systems on the market before 2 August get until 2 December 2026 to mark outputs. Older content needs no retroactive label.
Written by Sume