Subtitle burn-in

A closed caption is a separate text track the viewer can turn on. A burned-in subtitle is pixels. Short-form video runs on burned-in captions almost exclusively, for one reason: a caption the viewer has to enable is a caption most viewers never see.

Why the format demands it

Many people scroll with the sound off, at least until something makes them turn it on. A faceless video whose entire content is a voiceover is silent and meaningless to that viewer unless the words are on screen. Burned-in captions also survive being downloaded, reposted and cross-posted, which a caption track does not.

Styling that works at phone size

• Large. Big enough to read at arm's length on a 6-inch screen.
• High contrast. White text with a black stroke or a solid box, never thin text on a busy background.
• Word by word, or three words at a time. Long lines force reading instead of glancing.
• Centre of the frame. Not the lower third. See safe zones.
• Timed to the audio, not to an even interval, so the highlighted word matches the spoken word.

The trade-off

Burned-in text cannot be translated or corrected later. A typo means a re-render, and a video for another language needs a separate export. That is the price of being visible to the silent majority of the feed.

MakeViral builds captions from the exact character timings returned by the voice model rather than estimating them, so the highlight lands on the spoken word. See auto captions and the AI voiceover video generator.