School announcement video in two languages for parents: captions
One clean announcement clip, two caption jobs: English and your second language. Language hints, Korean styles and a human check on the translation.

To send a school announcement video to families who read different languages, keep one clean clip and burn two caption versions, one per language, as two Video captions jobs. Each job is $0.20 for a video up to 60 seconds (Sume docs, read 2026-10-03), so two languages cost $0.40 for the captions. Families get the file in their language, not a clip with two stacked lines of text. Sume does the burning; the translation is yours, and a person who reads both languages should check it.
Two ways to supply the words
If the clip has speech, a caption job can transcribe it, and language is a speech-to-text hint (en, ko and so on). If you translate the text yourself, send the translated lines as cues with text, start and end. Cues skip transcription, so timing comes from you; set them against the original clip.
| Version | Input | Style rule | Cost |
|---|---|---|---|
| English | Clean clip, speech transcribed or English cues | Omitted style picks slam for Latin text | $0.20 |
| Second language, Latin script | Clean clip plus translated cues | Omitted style picks slam | $0.20 |
| Korean | Clean clip plus Korean cues | Omitted style picks black-outline; slam is rejected for Hangul | $0.20 |
Language does not pick the style
language never selects the style or the font. An omitted style follows the wording: Korean copy resolves to black-outline, and naming a Latin style such as slam with Korean text returns 400 caption_hangul_text_latin_style rather than drawing empty boxes. For Hangul you can also pick a font. Other scripts are covered in what Sume documents for Japanese, Chinese and Arabic captions.
Before it goes home
Have the translation checked by a fluent reader, especially dates, times and room numbers, which a caption faithfully reproduces whether right or wrong. Do not put student faces in a clip without the permissions your school requires. Run the localized caption QA checklist on each version.
Sources
Related posts
More in Use cases
- Seedance 2.5 or Avatar video for a 30-second product ad?
Sume's Video Router lists seedance-2.5 for 4-30 s clips; Avatar Video plans 4-60 s of a talking presenter. How to choose for a product ad.
- Shopify 3D model or 360 spin video: what Sume can make
Shopify's 3D slot needs a GLB or USDZ file. Sume makes video, not 3D models. Shopify's limits for each, when a spin clip is enough, and the request to make one.
- Did the Shorts originality update cut my reach? What Studio shows
YouTube gave no dashboard for the Shorts originality change. Compare Shown in feed, Viewed vs swiped away and Shorts feed traffic for a before and after.
- One garment on different body types with AI: use real people
Shoppers want clothes on bodies like theirs. Google's 2023 try-on showed real models XXS to 4XL. Build a range from consented photos and one Sume image call.
Written by Sume