Open vs closed captions — and where subtitles and SDH fit in

Four terms, defined

People use these words interchangeably, but they describe different things — and the difference decides what your viewers can do.

• Open captions are part of the picture. They were rendered into the video frames at export, so nobody can turn them off. You’ll also see them called burned-in captions, hardcoded subtitles or hard subs.
• Closed captions are a separate track of timed text — a sidecar file such as SRT or VTT, or caption data carried with the video — that the player draws on screen only when the viewer switches it on. The CC button is the giveaway.
• Subtitles are, in the HTML standard’s words, a transcription or translation of the dialogue for when the sound is available but not understood — typically a viewer watching in a language they don’t speak.
• SDH, subtitles for the deaf and hard of hearing, are subtitles written to do a caption’s job: dialogue plus speaker identification, sound effects and music. The term appears in the European Accessibility Act and in streaming style guides such as Netflix’s.

WCAG, the web accessibility standard, defines captions as a synchronized visual or text alternative for both speech and non-speech audio needed to understand the content. Its notes draw the same line as this guide: closed captions can be turned on and off with some players, while open captions can’t — for example, when the text is embedded in the video image. The notes also point out that in some countries captions are called subtitles. The BBC’s guidelines, for instance, use “subtitles” for text primarily intended for viewers with hearing loss.

Open captions (burned in)

Open captions are pixels. Once the video is exported, the words are as much a part of the frame as the speaker’s face. That cuts both ways.

What you gain

• Complete control of the look. Font, size, color, outline, animation and position are exactly what you designed. Word-by-word highlights and brand styling are only possible this way.
• The same result everywhere. The captions survive downloads, reposts, messaging apps and platforms that don’t accept caption files, and they don’t depend on anyone’s settings.
• Always on. Nobody has to find a CC button. In a feed where people scroll with the sound off, that’s the point.

What you give up

• Viewer control. Someone who needs larger text or higher contrast can’t change it, and someone who finds captions distracting can’t hide them.
• Machine-readable text. To software, burned-in captions are part of the image. Screen readers and braille displays can’t read them, and a player can’t translate them or show them in another language.
• Easy fixes. A typo means exporting and uploading the video again.
• Multiple languages. Each language needs its own export.
• A clear frame. Burned-in text can collide with on-screen graphics, app buttons, or a platform’s own captions if a viewer switches those on.

Closed captions

A closed caption track is text with timings. The player decides how to draw it, and the viewer decides whether to see it.

What you gain

• Viewer control. Captions can be switched on and off, and many players let viewers change their size, color and background. In the US, FCC rules require covered video devices — and the apps distributors use to deliver TV programming online — to let viewers change caption text size from 50% to 200% of the default and choose text color, opacity and fonts. The European standard EN 301 549 likewise asks technology that displays captions to let users adapt how they look.
• Real text. The words can be edited, searched, translated into more tracks and processed by assistive technology. EN 301 549 even describes a mode that speaks captions aloud.
• Easy to fix and extend. Correct a word by replacing the file. Add Spanish or German as extra tracks on the same video.

What you give up

• They’re off until someone turns them on, and not every viewer will.
• The player decides the look. Styling is limited. YouTube, for example, ignores styling in SRT files and supports only bold, italic and underline in VTT files.
• They don’t travel. Not every app or upload flow accepts a caption file, and a caption track doesn’t come along when a video is downloaded and reposted.

Subtitles and SDH

Subtitles and captions can each be open or closed. The difference is who they’re written for. Subtitles assume the viewer can hear, so they carry the dialogue and little else. Captions and SDH assume the viewer can’t, so they also carry what the soundtrack tells you. The same moment, both ways:

12
00:03:15,200 --> 00:03:16,900
Did you hear that?

12
00:03:14,300 --> 00:03:15,100
[door slams]

13
00:03:15,200 --> 00:03:16,900
[Maya] Did you hear that?

Style guides format this differently. Netflix’s English guide puts speaker IDs and sound effects in brackets, in lowercase except for proper nouns, and uses them only when the viewer can’t see who or what it is. The BBC tells speakers apart with text color and writes sound effects in white capitals. What matters is that viewers who can’t hear still get the information. If a player lists both “English” and “English SDH”, the SDH track is the one with the sound information.

Pros and cons side by side

What to use where

TikTok, Instagram Reels and YouTube Shorts

Burn them in. Vertical short-form video plays in a feed on a small screen, often muted, with the app’s interface over the bottom and right edge of the frame. Burned-in captions let you size the text for a phone, keep it clear of the buttons, and use the word-by-word styles that suit the format. Some apps generate their own automatic captions too, but you don’t control their accuracy, their look or whether they’re on, and if a viewer turns them on, the two sets of text can overlap. See how to burn subtitles into a video, and caption best practices for size and placement.

YouTube and Vimeo

Upload a closed caption file. Both platforms accept SRT and VTT, and viewers can switch captions on when they need them. YouTube also creates automatic captions, but its help center warns they can misrepresent speech because of accents, mispronunciations or background noise, so a checked file is better. The steps are in how to upload an SRT file to YouTube. If you also publish short cutdowns of a long video, burn captions into the clips and keep the caption file for the full version.

LinkedIn

For a talking-head clip in a professional feed, burned-in captions are the dependable way to be read with the sound off. LinkedIn also lets you attach an SRT file when you post from the desktop site, which gives viewers a closed track. If a video has both and a viewer turns on CC, they’ll see the words twice, so decide which matters more for that post.

Courses, training and company websites

Use closed captions: a VTT file on your own player, or the caption feature of your course or video platform. Viewers here may rely on captions to do their job or finish a course, and are more likely to need to enlarge or restyle the text. It’s also where accessibility rules are more likely to apply — see captions and accessibility law — and a closed caption generator will get you the file.

The accessibility trade-offs

Are burned-in captions accessible? They’re far better than no captions: a deaf viewer can follow the dialogue, and WCAG’s definition of captions includes open captions. But they fall short in ways closed captions don’t:

• Size and contrast are fixed. A viewer with low vision can’t enlarge the text or add a background. EN 301 549’s requirement to let users adapt caption appearance excludes captions shown as unmodifiable characters, and its notes give bitmap subtitles as an example. Burned-in captions are the extreme case.
• Assistive technology can’t read them. People who use screen readers or braille displays — including deafblind viewers — get nothing from text inside the image.
• They can’t be switched off, and they can cover important parts of the picture, which WCAG’s notes say captions shouldn’t do.
• Styling can work against reading. Fast word-by-word animation, thin fonts and weak contrast make captions harder for everyone. The best-practices guide covers the limits.

Where accessibility matters — a public body’s website, a course, a product page — publish a closed caption track, whether or not the video also has captions burned in.

Why many creators do both

Because the two do different jobs. Burned-in captions win attention in a muted feed and look the same everywhere. A closed caption file gives viewers control, gives the platform real text, and can carry several languages. Often the two end up on different versions of the same content: burned-in captions on the vertical clips, a caption file on the full-length video and on the website.

Doing both is cheap when both come from one transcript. Correct the words once, export a video with the captions burned in, and export an SRT or VTT file from the same edit, so a fix shows up everywhere and the versions never disagree. That’s the workflow AutoCap is built around: every word has its own timing, you fix what the transcript got wrong, choose a caption style, and export a burned-in MP4 — plus SRT and VTT files on paid plans.

If a single video carries both, some viewers will see the words twice when they turn on CC. For a social clip that’s usually a fair trade. For training or public-sector video, consider leaving the burned-in captions off so the closed track can do its job.

Questions

Are burned-in captions the same as open captions?

Yes. Burned-in captions, hardcoded subtitles and open captions all describe text rendered into the video frames. WCAG’s glossary describes open captions as captions that cannot be turned off.

Do closed captions only help deaf viewers?

No. They help anyone who can’t or doesn’t want to listen: people watching with the sound off, in a noisy place or in a second language, or following a fast speaker or a poor recording.

Is SDH the same as closed captions?

They carry the same information for the same audience: dialogue, speaker identification and important sounds. SDH, subtitles for the deaf and hard of hearing, is the term used in the European Accessibility Act and in streaming style guides; closed captions is the usual US term for a caption track viewers can switch on.