Burned-in captions vs SRT: what they are, when to use each

Burned-in captions, also called open captions or hardcoded subtitles, are text drawn into every frame of a video when it is exported. They show on any player or app, and viewers cannot turn them off. Closed captions come as a separate .srt or .vtt subtitle file, which the player reads and the viewer can switch on or off.

Below: how the 2 compare, whether burned-in captions count as accessible, why short-form video uses them, and how to make them, including from a subtitle file you already have.

What burned-in captions are

Burned-in captions are part of the picture. On export, the editor paints the words onto each frame, like a title or a sticker, so the result is one ordinary MP4. They go by 3 names:

• Burned-in captions: the usual term in video editors and caption tools.
• Open captions: the accessibility term. The W3C defines them as "captions that cannot be turned off" (WCAG 2.2, Understanding 1.2.2).
• Hardcoded subtitles, or hardsubs: the usual term when the text is film or series subtitles.

Closed captions travel as text beside the video. The most common format is an .srt file: plain text with a number, a start and end time and the words for each line.

1
00:00:01,000 --> 00:00:03,200
Burned-in captions are part of the picture.

A .vtt file (WebVTT) holds the same in the format web players read. Either way, the player draws the text and decides how it looks.

Burned-in vs closed captions

Burned-in captions win wherever the viewer has to see the words without doing anything. A subtitle file wins wherever the viewer, a screen reader or the platform needs the words as text.

Short answer: burn captions in for TikTok, Instagram Reels and YouTube Shorts. Upload a subtitle file for long YouTube videos and course lessons. When you post both a short cut and a long version, make both from the same transcript.

Are burned-in captions accessible

Burned-in captions meet the main web accessibility standard. WCAG 2.2 success criterion 1.2.2, at level A, asks for captions on all prerecorded video with sound, and it lists open captions (technique G93) as a sufficient way to meet it, next to closed captions (G87) (W3C).

Burned-in captions still leave some viewers out. The Bureau of Internet Accessibility lists 3 limits: viewers cannot change the size of burned-in text, screen readers find nothing to read because the text is part of an image, and the captions cannot easily be translated (BOIA, 2021). Large, high-contrast text helps with the first. For long videos, a subtitle file alongside the burned-in version covers the rest.

In WCAG, captions also name who is speaking and sounds that matter, such as laughter, so add those where the clip depends on them.

US caption rules for online video are narrow. FCC rules cover programs shown on US television with captions and later put online, including clips on the broadcaster's own site or app, not video made only for the internet (FCC).

Why burned-in captions suit TikTok, Reels and Shorts

Burned-in captions are the default for short-form video because they are the only captions every viewer is sure to see.

• Muted viewing: many people watch with the sound off. A 2019 survey of 5,616 US adults by Verizon Media and Publicis found 69% watch video with the sound off in public (reported by StreamingMedia). Text in the frame reaches them before they decide to scroll.
• Nowhere to put a file: YouTube, Facebook, LinkedIn and X take an uploaded .srt (Sprout Social). TikTok and Instagram offer their own auto captions instead, which viewers can switch off and you cannot style.
• Style: an .srt file carries no font, color or word highlight. If the look is part of the video, it has to be in the picture.
• One file everywhere: the captions travel inside the MP4, so a download, a repost or the same clip posted to 3 apps keeps them.

The catch is the platform's own interface. TikTok's buttons and post text sit over the bottom and right of the frame, so place the captions higher, as the TikTok captions guide shows. If you burn captions in, also check the platform's auto caption setting, so viewers do not see the words twice.

How to burn captions into a video

Burning captions in takes a timed transcript, a style and an export that draws the text onto every frame. Full disclosure: I build ReelCaption, so the steps below use it.

Burn captions in from a transcript

1. Open the ReelCaption studio in Chrome, Edge or Safari, which can export MP4. Other browsers can still download .srt and .vtt files.
2. Drop in the video. Whisper transcribes the speech and gives every word its own start and end time.
3. Read the captions on the timeline and fix names, slang and anything misheard.
4. Pick a font, colors, outline and a word highlight, and move the captions clear of the platform's buttons.
5. Export the MP4. Your browser encodes it at the original resolution, with no watermark on any plan.

The video stays on your device. Only the audio track, about 1 MB a minute, goes to Groq, which runs Whisper, and it is not stored. Your first 2 clips of up to 60 seconds need no account, then a Google sign-in gives you 5 free minutes a month.

Burn in your own .srt or .vtt file

If a transcription service or your editor gave you a subtitle file, add it next to the video in the studio. ReelCaption reads the file in your browser and skips transcription, so it uses none of your minutes or free clips. Then you style the captions and export the MP4 as above. The burn subtitles into a video page walks through it.

A subtitle file only says when each line starts and ends, not when each word is spoken. ReelCaption spreads each line's time across its words by their length, so the word highlight lands close to the speech rather than exactly on it. Retime any caption that drifts on the timeline.

Burn captions in with ffmpeg

If you have ffmpeg, one line burns an .srt file in:

ffmpeg -i video.mp4 -vf subtitles=captions.srt -c:a copy captioned.mp4

The subtitles filter needs an ffmpeg built with libass, and it re-encodes the video (FFmpeg wiki). There is no word highlight.

Common questions

Can you remove burned-in captions

No, not cleanly. Burned-in captions replace the pixels under them. You can crop or blur that part of the frame, which loses the picture there too, so keep the uncaptioned original.

Are burned-in captions the same as subtitles

Mostly, in everyday use. Strictly, captions are for viewers who cannot hear the audio, so they also name speakers and meaningful sounds, while subtitles carry only the dialogue. The W3C notes that some countries call captions subtitles.

Do burned-in captions meet WCAG and the ADA

For WCAG, yes. Open captions are a sufficient technique for success criterion 1.2.2. I found no ADA rule that names a caption format. The Justice Department's 2024 rule for state and local governments sets WCAG 2.1 level AA as the standard for their websites and apps (ADA.gov). If your organization has legal duties, ask someone who advises on them.

Do burned-in captions help SEO

Not directly. Search engines and platforms index the text you give them: the title, the description and an uploaded subtitle file. Burned-in words are part of the image, so put your key words in the post text, and on YouTube upload the .srt too.

Make both from one transcript

ReelCaption makes both from one transcript: an MP4 with the captions burned in, and .srt and .vtt files for YouTube or a course platform. It does not translate, so for subtitles in other languages you need a different tool. Try it on your next clip with the free caption generator.