Tire shop winter swap promo: voiceover plus timeline for about $1.27
A tire shop can build a 45-second winter-swap promo from three short clips, a text-to-speech voiceover and one timeline render. Itemized to the cent.

A tire shop can produce a 45-second winter-swap promo on Sume for about $1.27: three 6-second Wan 3.0 clips at 480p, a text-to-speech voiceover, and one Timeline 1.0 render. The render fee is $0.10 because 45 seconds rounds up to one minute.
Winter tire swaps cluster around the first cold snap, so the useful asset is a spot you can re-cut quickly when the forecast changes. This post lays out the pieces, the price of each, and the order to run them.
The three parts
First, picture. Generate three short clips with Wan 3.0 at 480p: a tire close-up on a lift, a technician torquing a wheel, and a car pulling out on a wet road. Wan 3.0 accepts 2 to 30 seconds, so each can be 6 seconds. See the Video Generation docs.
Second, voice. Sume's text-to-speech is priced at $0.0475 per 1,000 characters, up to 20,000 characters, using a voice from an avatar or a voice id. A 700-character script is 700 / 1000 x $0.0475 = $0.033, about 3 cents.
Third, assembly. Timeline 1.0 takes one audio spine and ordered video slots and renders a single MP4. The Timeline docs say the render charge reserves ceil(duration in minutes) at $0.10 each, and that there is a free plan endpoint to check a document first.
Itemized cost
Wan 3.0 at 480p costs $0.0625 per second on Sume, so a 6-second clip is $0.375, charged as 38 cents per job (read 2026-10-10). Three clips are $1.14.
Add 3 cents of voiceover and the $0.10 render and the promo is $1.27.
| Item | Basis | Price |
|---|---|---|
| Three Wan 3.0 clips | 3 x 6 s x $0.0625, 38 cents each | $1.14 |
| Voiceover | 700 characters at $0.0475 per 1,000 | $0.03 |
| Timeline render | ceil(45 s / 60) = 1 minute | $0.10 |
| Total | $1.27 |
Why the 60-second line matters
The render charge is per whole output minute. A 45-second spot costs $0.10. A 61-second spot reserves two minutes and costs $0.20. If your voiceover runs a few seconds long, tighten the script rather than paying for a second minute. The existing post on the ceil-minute trap walks through it.
Script and music guidance
Keep the script to one offer and one call to action: the swap window, a booking line and the address. Read it aloud once before you generate; at a normal pace 700 characters is a bit under a minute, so trim it until it lands near 45 seconds.
Timeline supports a soundtrack with ducking under the voice, which needs an audio spine to duck against. If you want a bed, generate it with the Music Router at $0.125 per track and keep it quiet. That adds $0.125 and takes the total to about $1.40.
Do not name specific tire brands or quote prices a vendor would have to approve. Say the shop's own offer.
Shot list and timing
A 45-second spot with three clips needs more than three 6-second shots, so plan the timeline before you generate. The audio spine sets the length, and each video slot can be trimmed or looped to fit. Think in beats: a 10-second opening that names the problem (the first frost), a 25-second middle that shows the swap and the offer, and a 10-second close with the address and the phone number.
If you want the visuals to cover 45 seconds with only three 6-second clips, reuse them. Use the free plan endpoint first to confirm the document is valid and to see the cost, then submit the render. Generate extra clips only if the reuse looks repetitive. Each extra 6-second clip at 480p adds 38 cents.
Keep the claims modest and local. A shop can say it is open Saturdays and takes appointments. It should not claim a stopping distance or a safety result unless it can back that with something it owns.
Because each clip and the voiceover are separate files, a second version costs only what you change: a new script at 3 cents and a new render at $0.10. Keep the clips and re-run the timeline with a different voiceover when the forecast moves the swap date.
Sources
Related posts
More in Use cases
- Toast Websites video: 50 MB, landscape only, uploaded by hand
Toast Websites takes a brand video up to 50 MB in landscape only and does not pull it automatically. Make a 16:9 holiday-menu clip with Sume and trim it to fit.
- Tour Operator Trip Teaser: 30 Seconds in One Job, and Kling 4.0?
Kling 4.0 is listed with 3 to 30 second clips, but Sume lists Kling 3 at 15 s. For a 30-second teaser in one job, Sume offers Wan 3.0 and Seedance 2.5.
- Twelve colourways times three takes: n=3 versus 36 calls on Sume
For 12 colourways with 3 takes each, use 12 calls with n=3, not 36 calls with n=1. Same cost, fewer requests. Totals on FLUX.2 Pro and Seedream 4.5 inside.
- India's upload prompt: declare AI video before a big platform shows it
Under India's IT Rules, large social platforms must ask uploaders to declare synthetic media and verify it. What to answer, and what to log per Sume job.
Written by Sume