Baby shower video music: a gentle 90-second slideshow bed

For a baby shower slideshow, ask for a gentle music-box and acoustic-guitar lullaby of about 90 seconds with a soft swell where the photo of the parents appears. On Sume that is one generation at $0.125 and a render that bills as two started minutes at $0.20, so $0.325.

Slideshows are paced by the photos, so the music has to be patient. The brief asks for a steady, sweet, unfussy piece and one soft lift. It does not ask for lyrics or names; those belong in the on-screen text.

The brief to send

The structure below places the lift at 0:45, which is where most slideshows move from the bump photos to the announcement photos.

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: baby-shower-video-music-a-gentle-90-seco-001" \
  -d '{
      "model": "sume/music-auto",
      "prompt": "Sweet lullaby for a baby shower, 80 BPM, G major. Music box, soft acoustic guitar, light glockenspiel, warm strings entering at 0:45. Tender and joyful, gentle swell at 0:45, soft close at 1:30. Warm, close, storybook production. Instrumental, no vocals. A 90-second track."
    }'

Why each axis is set that way

Getting the length right

The router has no length field, so 90 seconds go in the prompt. Set audio.duration_seconds to 90 in a silent-spine render, with the track as soundtrack.url and gain_db for level. With no voice, ducking is not needed and is refused on silence anyway.

If the host adds a spoken toast, use that recording as the spine instead, and set duck_db so the lullaby dips under it.

Before paying for a render, call POST /v1/timeline-1.0/plan with the same body. It is unbilled and reports the billable minutes, which for this 90-second video is 2. Then send the real POST /v1/timeline-1.0/render with an Idempotency-Key, wait up to 30 seconds in sync mode, and poll GET /v1/jobs/:id/status if it is still queued or processing. Statuses are queued, processing, completed, failed and canceled, and a 429 queue_full means try again shortly.

What it costs

Ninety seconds bills as two started minutes. Trimming the slideshow to 60 seconds would halve the render to $0.10. For a family with three events, such as the shower, a gender reveal and a first birthday, three generations and three renders come to $0.975. Scaling is simple arithmetic. One finished video here is $0.325 of audio and assembly, so ten of them are $3.25 and fifty are $16.25, before any retries. Failed jobs are not billed as accepted generations, but a regeneration you ask for is a new job, so budget one or two extra takes for a hero clip.

Check before you publish

Because a generation is cheap and fixed, you can offer two moods, lullaby and upbeat, and let the host choose.

Keep the prompt and the job id so you can reproduce the track.

The hand-off between the two calls is the step people miss. When the music job completes, GET /v1/jobs/:id/result returns result.artifacts[] with an audio/mpeg file on media.sume.com. Timeline only accepts workspace media URLs, so import any outside file through POST /v1/media-imports first, and send the resulting URL as soundtrack.url. The Music docs also note that job.request.routed_model names the engine that actually ran, which is worth saving with the file.

• Use the family's photos only with permission from everyone shown.
• Keep the music soft enough for a living-room speaker.
• Review the whole file for any accidental vocal sounds.
• Confirm usage terms for generated music with Sume before posting beyond private sharing.