Wedding venue teaser music: write the prompt with section markers
Score a 30-second venue teaser with Sume's Music Router: no duration field, so length goes in the prompt with timed section markers. Example prompt and request.

To score a 30-second wedding venue teaser, send a prompt to POST /v1/music-router/generate and put the length in the prompt. Sume's Music Router rejects duration and duration_seconds, and the docs tell you to control length with wording like "a 30-second track" or section markers such as [0:00-0:10] Intro: ....
The default model id is sume/music-auto, which picks the engine (Lyria 3.5 today). You can also pin lyria-3.5 or lyria-3-pro from the router catalog.
Why a planner needs a custom bed
A venue teaser is a few clips of the same place: the arrival, the ceremony lawn, the reception hall. Music is what makes the cuts feel intentional, and a generic stock bed repeats across every planner in town. A prompt written for your three scenes gives you a track that rises where your cut rises.
The music prompt accepts 1 to 5000 characters. Put exclusions in the positive text: say "no vocals, no spoken word" instead of using negative_prompt, because Sume does not support a non-empty negative prompt.
Write the prompt in three timed sections
Name tempo, instrument and mood for each section, and keep the sections tied to your shot list. Sume's music docs suggest a scene-specific brief rather than a generic genre label, and note that the axes you write are directions, not guaranteed outputs, so listen to the result.
An optional image_url can condition the music on a venue photo.
curl -X POST https://api.sume.com/v1/music-router/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: venue-teaser-music-001" \
-d '{
"model": "sume/music-auto",
"prompt": "A 30-second instrumental, no vocals, no spoken word. [0:00-0:10] Arrival: solo acoustic guitar, warm and slow. [0:10-0:22] Ceremony: strings enter, gentle build. [0:22-0:30] Reception: light percussion, bright resolve, clean ending."
}'Fetch the audio and cut to it
Poll /v1/jobs/{id}/status, then read the audio artifact from result.artifacts[] where type is audio. The exact length can differ from what you asked for, so measure the file before you cut picture to it.
To assemble the teaser, Sume's Timeline 1.0 takes one audio spine plus ordered video slots and returns one MP4. Declare audio.duration_seconds to match the track you got, and trim your clips to fit.
| Item | Value | Note |
|---|---|---|
| Endpoint | POST /v1/music-router/generate | Async job |
| Prompt | 1 to 5000 characters | Put exclusions in the positive text |
| Length control | Prompt wording or section markers | duration is rejected |
| Model ids | sume/music-auto, lyria-3.5, lyria-3-pro | Auto is the default |
| Pricing | Fixed Music price per audio generation | See the pricing page for the live number |
Rights and taste
Generated music is not a recording of a famous song, and you should not ask for one by name. If a couple wants their own song under the final film, license it separately and use Sume's timeline to lay it in.
Sources
Related posts
More in Use cases
- Wedding save-the-date video: photo to end card with Gemini Omni
Make an 8-second save-the-date by pinning a couple photo as first frame and a designed date card as last frame in Gemini Omni on Sume. Request and cost.
- Save-the-date video from two photos: Wan 3.0 first and last frame
A 5-second save-the-date that travels between two photos costs 63 cents at 720p on Sume's wan-3.0. How frame_images works and what to avoid.
- Weekly toolbox talk video with an AI avatar: 52 clips a year, cost
A weekly 30-second avatar safety reminder costs a fixed sum per year at each tier. Script rules, review steps, and what it must not replace.
- Which Google image model for product photos after Oct 6?
Nano Banana 2.1, Pro or Lite for product photos? A table built from Google's Oct 6 prices, size limits and reference rules, and what Sume lists.
Written by Sume