A 3-minute Short with a ducked music bed: $0.425 on Sume
YouTube Audio Library songs cap at 90 seconds in a 3-minute Short. Generate one Music 1.0 track for $0.125 and render with ducking for $0.30.

A 180-second Short with a music bed under your voice costs $0.425 on Sume: Music 1.0 is a fixed $0.125 per generation and a 180-second Timeline 1.0 render is 3 minutes at $0.10 = $0.30 (read 2026-10-08). The reason to generate a track is that YouTube's Audio Library songs may be used for only up to 90 seconds in a 3-minute Short, and some are limited to 60 or 30 seconds.
The YouTube constraint
YouTube says a Short can run up to three minutes in a square or vertical frame. The same page says Shorts Audio Library songs are usable up to 90 seconds in a 3-minute Short, some limited to 60 or 30 seconds, and a Short over one minute with an active Content ID claim is blocked globally and loses monetization. A track you generate and mix yourself is not an Audio Library song, but Content ID risk is yours to judge; Sume does not clear rights for you.
The cost, line by line
Music 1.0 resolves through the Music Router, today to Lyria 3.5, and charges a fixed $0.125 for each accepted generation. It has no duration parameter, so steer length in the prompt, for example 'a 3-minute track'. The Timeline render reserves ceil(180 / 60) = 3 minutes.
| Line | Arithmetic | Cost |
|---|---|---|
| Music 1.0 generation | Fixed per generation | $0.125 |
| Timeline 1.0 render | 3 minutes x $0.10 | $0.300 |
| Total | $0.125 + $0.300 | $0.425 |
Ducking and fades
The soundtrack block takes url, gain_db, loop, fade_out_seconds up to 10, and duck_db from 0 to 20. Ducking needs a real audio spine, so it works with your narration but not with audio.mode: silence. If the generated track comes back shorter than 180 seconds, set loop: true. Check the returned length rather than assuming it, since the docs promise only full-length songs of up to a few minutes.
{
"audio": { "url": "https://media.sume.com/artifacts/artf_demo/narration.wav", "duration_seconds": 180 },
"video": [
{ "source_url": "https://media.sume.com/artifacts/artf_demo/part1.mp4", "start": 0, "duration": 90 },
{ "source_url": "https://media.sume.com/artifacts/artf_demo/part2.mp4", "start": 90, "duration": 90,
"transition": { "type": "dissolve", "duration": 0.5 } }
],
"soundtrack": {
"url": "https://media.sume.com/artifacts/artf_demo/bed.mp3",
"gain_db": -6, "loop": true, "duck_db": 12, "fade_out_seconds": 4
},
"output": { "width": 1080, "height": 1920 }
}Prompt tips for the bed
Exclusions go in the positive prompt because negative_prompt is not supported. Write 'instrumental, no vocals, no spoken word, a 3-minute track' rather than a style name alone. Keep the mix at a level that works when ducked, and listen to the first 10 seconds before you spend the render.
Sources
Related posts
More in Use cases
- 30 jingle candidates for 5 brands: $3.75, then trim 5 to 15 seconds
Six Music 1.0 candidates for each of five brands is 30 x $0.125 = $3.75. Trimming five winners to 15 seconds in Timeline adds $0.50. The arithmetic.
- 30-second ad: voice, music, captions and render cost 45.5 cents
A 30-second ad's finishing stack on Sume is 45.5 cents: 3 cents of voice, a $0.125 music track, $0.20 captions and a $0.10 render, before footage.
- A 30-second launch hero: Seedance 2.5 $17.34, Wan 3.0 $3.75
Priced at 720p on Sume: a 30-second hero in one Seedance 2.5 job, one Wan 3.0 job, or three 10-second Omni Flash clips joined. The gap is large.
- Turn a 30-second voice memo into a vertical status video
Put a 30-second voice recording under four stills and burned captions for a WhatsApp status or any vertical feed. Timeline body and a total of about 33 cents.
Written by Sume