Burned-In Captions vs Closed Captions: Which to Use

Every captioned video makes one choice before anything else: are the words part of the picture, or a separate track the viewer can switch off? That is the whole of burned-in captions vs closed captions, and getting it right depends less on taste than on where the video is going. This post explains the difference, what each kind is good at, which to use where, and how to make burned-in captions without sending your footage anywhere.

The difference, in two sentences

The W3C's accessibility guidance on captions and subtitles puts it plainly. Most captions are closed captions, which viewers can show or hide; open captions are always displayed and cannot be turned off.

Burned-in captions are open captions. The words are drawn into the frames of the video when it is exported, so they are part of the picture, exactly like a title. Closed captions live outside the picture, usually in a text file such as WebVTT, which MDN describes as text cues that are time-aligned with the video, each with a start time, an end time and its text. The player reads that file and draws the words over the video when the viewer asks for them.

You will also see the same pair called hardcoded and soft subtitles. The idea is identical.

What burned-in captions are good at

• They go wherever the file goes. Repost the video, send it in a message, drop it into another platform: the captions are in the picture, so they travel with it. Nothing depends on a player supporting caption files.
• You control how they look. Typeface, size, colour, position and how many words appear at once are all your decisions, not a player's defaults.
• They are on for everyone. In a feed where many people scroll with the sound off, nobody has to find a button for them to appear.

The costs are the mirror image. A viewer cannot turn them off, cannot make them larger, and cannot switch them to another language. The text is pixels, so a player has no way to read it as text.

What closed captions are good at

• The viewer decides. People who want captions turn them on; people who find them distracting leave them off.
• More than one language. Because the captions are separate text, one video can carry several tracks. YouTube, for example, lets creators add subtitles in additional languages from YouTube Studio.
• They stay text. A caption track is text the player can work with, rather than part of the image.

The cost is dependence. Closed captions only appear where the player supports them and the viewer enables them, and they look however that player draws them.

Which to use where

There is no universal answer, but there is a sensible default for each kind of destination.

• Short-form feeds and anywhere the video autoplays muted: burned-in. The captions need to be there in the first second, styled to stay clear of the platform's buttons, and our posts on captions for Instagram Reels and editing auto-generated captions cover the details.
• Long-form video on a platform with a proper caption system, such as YouTube: closed captions are usually the better fit, because viewers can switch them off and you can add languages.
• A video you will post to several places: burned-in, so one correction pass covers every copy.
• Never both at once. Burned-in captions plus a platform's automatic captions shows the viewer the same words twice. If you burn them in, leave the platform's captions off for that post.

Making burned-in captions in flshy

flshy makes burned-in captions: every cue is a text clip on the timeline, and the export draws them into the picture. It runs free in flshy, in the browser, and transcription happens on your own machine. The full walkthrough is add captions to a video.

1. Cut the clip first. Caption cues sit on their own track and do not move with a silence removal pass, so caption once the clip is at its final length.
2. Generate the captions. Select the clip, open Captions, pick the spoken language or leave it on auto detect, and press Generate captions. Transcription runs on your device, in 14 languages plus auto detect.
3. Put them on the timeline. Press the button under the transcript that reads Add 24 as Editable Text, or whatever your count is.
4. Correct and annotate. Each cue is an ordinary editable text clip. Fix names and jargon, and type in any sound that matters, such as [music] or [laughter].
5. Style and place them. Pick a preset, then use the Properties tab to move the captions up, clear of the platform's interface. Generated cues are five words at most, which keeps them readable.
6. Export. H.264 MP4 up to 4K, with the captions burned in and no watermark.

If you need closed captions instead

flshy's captions are drawn into the picture when you export, so when a platform's caption system is the better route, use that platform's own tools. On YouTube, automatic captions are generated in the video's default language, and creators can upload a subtitle file or add languages in YouTube Studio.

It works in the other direction too. If you already have an SRT or VTT file and want the words burned in, import it into flshy and every cue becomes an editable text clip. Imported cues arrive in plain styling, so restyle them before you export.