Singles Day 11.11 Chinese voice-over countdown: three clips for $1.98
A 15-second 11.11 countdown for cross-border sellers: three Wan 3.0 clips, a Chinese voice-over and one Timeline render, priced from docs at $1.98 on Sume.
A 15-second 11.11 countdown with a Chinese voice-over costs $1.98355 on Sume: three 5-second Wan 3.0 clips at $0.625 each, one Chinese TTS job of about 180 characters at $0.0086, and one Timeline render at $0.10. It uses no on-screen caption job, for a reason explained below.
Existing posts price Singles Day clip budgets and a 24-hour template. This one covers the voice: a Mandarin line over a short vertical montage, for a seller whose audience hears the offer in Chinese.
Three clips of 5 seconds
Take three product stills: the hero product, the bundle and the gift box. Animate each for 5 seconds with Wan 3.0 at 720p, $0.125 per second, so $0.625 each and $1.875 for three. The Video Router accepts 9:16 for Wan 3.0 and Seedance 2.5 (Video Router). If you pick Seedance 2.5 at 720p, 5 seconds is $2.889, so the three clips are $8.67 instead.
A Chinese voice-over
TTS 1.0 prices per character, $0.0475 per 1,000, with every character counted including punctuation. Chinese scripts are short in characters: a 15-second line of about 60 characters is $0.00285, and a three-line script of 180 characters is $0.00855. The voice library lists zh among its language codes; set language to zh on the job, as the TTS schema asks for every non-English transcript. generation_config.speed runs from 0.6 to 1.5 if the line needs trimming to the clip length.
Render, and where captions fit
One Timeline render joins the voice as the audio spine with three 5-second slots and a fade transition between them: $0.10 for 15 seconds (Timeline). The caption docs describe styles for Latin text and for Hangul, and do not document Chinese glyph coverage. I have not verified Chinese captions, so this build puts any Chinese offer text into the still before you animate it, rather than burning it afterwards. If you want captions, run one test clip and check the glyphs first (video captions).
| Line | Basis | USD |
|---|---|---|
| 3 x Wan 3.0 clip, 5 s, 720p | $0.125 per second | $1.875 |
| TTS, zh, 180 characters | $0.0475 per 1,000 | $0.00855 |
| Timeline render | $0.10 per output minute | $0.10 |
| Total | $1.98355 |
Rounded and variants
The total is $1.98355 before rounding; the wallet charges each job at its own rounding, so your three line items may sum a cent differently. A second voice, for a different promotional line, adds one TTS job and one render: about $0.11.
Related posts
More in Use cases
- Six-shot character consistency: Kling vs Sume references
Kling is described as keeping a character consistent across up to six shots. On Sume, use reference images or a reusable Avatar handle, and review every shot.
- Skippable YouTube ads: check the first 5 seconds with stills
Skippable in-stream ads can be skipped after 5 seconds and Google recommends under 3 minutes. Sample the pre-skip frames of a Sume clip with video inspect.
- Small-business holiday ad: three offers, two lengths for under $10
Three offer variants at 6 s and 10 s each, nine renders in all, cost $27.7344 on Seedance 2.5 and $6.00 on Wan 3.0 at 720p vertical on Sume.
- Snap ad video size: API says 1080x1920, ad specs page says 720x1280
Snap's Ads API lists 1080x1920 for top snap video while its ad specs page lists 720x1280. Render one 9:16 file at 1080p and check the pixels with Sume.
Written by Sume