A 60-second ad on Sume: voice, music, render, captions for 47 cents

The audio, assembly and caption side of a 60-second ad costs $0.46775 on Sume list prices, a little under 47 cents with reservations rounded up to cents: 900 characters of voice-over ($0.04275), one music track ($0.125), a one-minute Timeline 1.0 render ($0.10) and one caption job ($0.20). The video clips and stills you place in the timeline are extra and depend on the model.

Prices are from the Sume API catalog and the timeline and caption docs (read 2026-10-09). The 900-character script is an assumed size for about a minute of speech.

The line items

Voice-over: 900 characters x $0.0475 per 1,000 = $0.04275. Music: a flat $0.125 per generation, whatever the length. Timeline render: $0.10 per output minute, rounded up, minimum one minute, so a 60-second ad is exactly $0.10 (a 61-second ad would be two billable minutes, $0.20). Captions: $0.20 per job for videos up to 60 seconds.

The sum is $0.04275 + $0.125 + $0.10 + $0.20 = $0.46775. The catalog reserves in whole cents, so the held amount may round each line up; the arithmetic here is on list prices.

Where the money moves

Captions are the largest of the four lines, at 43 percent of the total. If your pipeline already has word timings from TTS, you can pass them as words and skip speech-to-text, but the caption job price does not change.

The render is the cheap part. A 60-second ad is the same $0.10 whether the clips are stills or footage, since the render uses worker ffmpeg only. Mind the 61-second cliff: trim the spine to 60 seconds to stay in one billable minute.

Order of operations

Generate the voice first, because its length sets audio.duration_seconds on the timeline. Add the music as the soundtrack with duck_db so the voice stays clear. Render, then caption the finished MP4. Use an Idempotency-Key on each paid submit and poll each job by id, as the jobs and results guide describes.

Variations

A 45-second ad with 700 characters costs $0.03325 + $0.125 + $0.10 + $0.20 = $0.45825. A 30-second spot with 450 characters costs $0.021375 + $0.125 + $0.10 + $0.20 = $0.446375. The render and caption lines are the same, so the voice is the smallest line at every length.

Three music candidates add $0.25 to any of these. If the caption job is the largest line, consider whether the ad needs on-screen text at all, or whether you can burn your own cues.

None of this includes the video itself. Clips and stills are priced per model and per second in the catalog, so add them to the table before you quote a total to anyone.