Graduation video music: a 2-minute anthem with a build

For a graduation video, ask for a 2-minute uplifting orchestral-pop anthem that builds to a triumphant lift at 1:20, where the diploma reel or the class photo lands. On Sume that is one generation at $0.125 and a two-minute render at $0.20, so $0.325.

A graduation video is emotional and has an obvious peak. The brief should therefore have a real arc: quiet start for the memories, rising middle, a clear climax and a warm resolution. Name the climax time, because that is the anchor your edit needs.

The brief to send

The sections below give a slow piano opening, a steady build with percussion, and the lift at 1:20.

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: graduation-video-music-a-2-minute-anthem-001" \
  -d '{
      "model": "sume/music-auto",
      "prompt": "Uplifting graduation anthem, 92 BPM, E-flat major. Piano, warm strings, snare build, French horns, light cymbal swells. [0:00-0:40] Gentle piano memories. [0:40-1:20] Steady build. [1:20-1:50] Triumphant lift. [1:50-2:00] Warm resolve. Cinematic pop production. Instrumental, no vocals. A 2-minute track."
    }'

Why each axis is set that way

Getting the length right

Section markers are the way to steer structure in Lyria 3.5. The router rejects duration and duration_seconds, so the two minutes are in the prompt. Check where the lift actually lands in the returned file and align the diploma reel to it.

Set audio.duration_seconds to 120 in the render and fade_out_seconds to 2 or 3 for a smooth finish. If a speaker's voice is part of the video, make that the spine and set duck_db.

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 120-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

A two-minute render is $0.20. Per-class versions, for example three schools with their own mood, are three generations at $0.375 plus three renders at $0.60, which is $0.975. The cost scales with the number of videos, not their length. 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

Seasonal spikes happen in May and June in many places; prepare the briefs early.

Keep the best-performing brief for next year's class.

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.

• Do not name the school or students in the music prompt; put names on screen.
• Check that the lift lines up with the reel before final export.
• Provide a version without music if the venue plays its own.
• Confirm usage terms for generated music with Sume before distributing.