Synthesia burned-in captions on dubbed videos, and Sume captions

On 9/30/2026 Synthesia's product updates page lists an improvement titled "Add burned-in captions to dubbed videos": captions can be added to a dubbed video in one click, so localized content stays readable with sound off. Burned-in means the text is drawn into the video frames, so viewers cannot switch it off.

If your dubbed video was made somewhere else and you only need the captions, Sume's standalone video captions job takes a public HTTPS video_url and returns a captioned video. This post sticks to what the Synthesia pages say and to what the Sume docs describe, read on 2026-10-02.

What did Synthesia actually announce?

The updates page gives a title, a one-sentence description and the date 9/30/2026, labelled "Improvement". It does not list a price, a language count for this feature or an API field, so none is claimed here.

Synthesia's help article (dated July 23, 2026 on the page) separates three caption types. Closed captions can be toggled by viewers and cannot be styled. Dynamic captions are styled on the canvas. Burned-in captions are permanently embedded, cannot be turned off by viewers, are not customizable, and are added at the generation step. The article also says the Generate panel's "Burned-in captions" toggle controls whether closed captions are in the exported file.

How does Synthesia's dubbing flow fit in?

Synthesia's dubbing docs describe AI Dubbing as translating a video's spoken content into another language, with an optional lip sync step, into 130+ languages. The batch section says you pick target languages, full dub or subtitles only, lip sync and video timing once, and the settings apply to every video in the batch.

The new burned-in option sits on top of that flow. The docs page and updates page I read do not say whether the burned-in text is styled like dynamic captions, so treat it as the plain, unstyled kind described in the help article until you see your own export.

How do I burn captions onto a finished video on Sume?

Create a caption job with the video URL. Sume's docs list style, font, language, script_text, design overrides, and authored cues as optional fields, and say language is only a speech-to-text hint: it never picks the style or font.

Speech captions need audible speech. A silent clip fails as caption_no_speech; pass cues or segments with text, start and end to burn authored text instead. Korean text on slam, punch or tiktok-green is rejected with caption_hangul_text_latin_style, so pick a Hangul style for Korean.

curl -X POST https://api.sume.com/v1/video-captions \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: dubbed-es-captions-001" \
  -d '{
    "video_url": "https://example.com/dubbed-es.mp4",
    "language": "es",
    "style": "slam",
    "mode": "async"
  }'

What should I check before choosing?

A short checklist, based only on the pages cited above.

• Do you need viewers to toggle captions? Then burned-in is the wrong type on either platform; use a caption file.
• Do you need to style the text? Synthesia's help article says burned-in captions are not customizable; Sume documents style and design overrides.
• Does the clip have speech? Sume needs it unless you supply cues.
• Is the video already dubbed? Sume's caption job works on any public HTTPS video URL; it does not dub.