Add Captions to a Video

Who captions are for

Subtitles assume the viewer can hear. They carry the dialogue, often in another language, and leave the soundtrack to do everything else. Captions assume the viewer can't hear, or has the sound off, so they have to carry the soundtrack too. That one difference changes what you write.

Deaf and hard-of-hearing viewers are the audience captions were designed for, and they notice at once when a laugh, an off-screen voice or a sudden silence goes unmentioned. A second, much larger group watches muted: on a train, in an open-plan office, in a feed that starts videos silently. A third switches captions on to follow strong accents, fast talkers or a language they are still learning. A caption track written for the first group serves the other two as well; the reverse isn't true.

In British English "subtitles" often covers both ideas, so a viewer in London may say subtitles where someone in Chicago says captions. The full distinction is laid out in captions vs subtitles.

Open captions or closed captions

The second decision is where the captions live.

Open captions

Drawn into the picture. Every viewer sees them in every player, styled exactly as you set them, and nobody can hide or resize them. In this editor, Render video produces open captions as an MP4.

Closed captions

Kept in a separate timed-text file that the player shows on request, usually behind a CC button. Viewers choose whether to see them, and some players let them change the size. Export SRT or Export VTT gives you that file.

Autoplaying social feeds, messaging apps and a presentation on someone else's laptop favour open captions, because there is no guarantee a caption file travels with the video. YouTube, Vimeo, course platforms and your own website player handle closed captions well, and there the viewer's ability to switch them off is a feature. You can also do both: render the open-caption MP4 for the feed and upload the SRT next to the clean original everywhere else. Open captions goes further into the burned-in route, and the closed caption generator is the page for producing a caption file on its own.

What a caption carries that a subtitle doesn't

Viewers who rely on captions recognise a handful of conventions:

• Speaker names when the speaker is off screen or not obvious, written before the line: "MAYA: We're live in five."
• Sound cues in square brackets for sounds that affect the meaning or the story: [phone buzzing], [audience laughs], [door slams]. Leave out sounds that change nothing.
• Music described by its role rather than its mere presence: [tense music], [upbeat pop song], or the lyrics themselves with a music note when the words matter.
• Manner of speaking when it changes what the words mean: [whispering], [sarcastically], [in Spanish].
• Silence when it is the point, such as [no audio] over a muted stretch, so nobody wonders what they missed.

Keep cues short and consistent. Choose one style for brackets and capital letters at the start and keep it across the whole video. Captioning sound effects and music covers the sound side in depth, and viewers of a full film or series usually expect the SDH subtitle conventions.

Captioning a video, start to finish

1. Open the editor and choose the video: MP4, MOV, M4V, WebM or MKV.
2. Get the dialogue in. Leave "Write the subtitles automatically with AI" on for a draft, or switch it off and use Import SRT / VTT or type each line.
3. Watch once with the sound up and correct the words. AI drafts contain speech only, so this is also the pass where you type speaker names and bracketed sound cues into the lines.
4. Where a sound needs its own caption, pause on it and press Enter to add a line at the playhead, then type the cue.
5. Clear the warnings in the line list, especially lines that became too fast to read once a cue was added.
6. Open Style, choose a legible preset such as High Contrast or YouTube, and check the position with the safe-area guide.
7. For open captions, click Render video and download the MP4. For closed captions, use Export SRT or Export VTT. For both, do both.

Reading speed decides whether captions work

Adding "[crowd cheering]" makes a line longer without giving it more screen time, and that is the most common way captions turn unreadable. The editor flags any line faster than 21 characters per second in Latin and most other scripts (8 for Japanese, 9 for Chinese, 12 for Korean), any line on screen for under 0.8 seconds, any line longer than 42 characters and anything over 3 lines. When a cue pushes a line past the limit, give the cue a line of its own just before or after the speech, stretch the line into a nearby gap, or trim filler from the dialogue. Captions don't have to be verbatim: dropping "you know" and false starts keeps the meaning and buys the reader time. Subtitle timing guidelines goes deeper on durations and gaps.

Contrast, size and position

Captions that depend on the picture behind them for contrast will fail on some frame. A box behind the text guarantees contrast everywhere: High Contrast puts yellow text on solid black, and the YouTube preset uses white on a dark box. If you'd rather have no box, keep a thick outline. Size is set relative to the video's shorter side, so a setting that reads well at 1080p scales with a 4K or vertical source. Keep captions off faces where you can, because some viewers lip-read alongside the text, and lift them above lower-thirds or platform buttons with the position and vertical offset controls. Subtitle backgrounds compares box, outline and shadow in practice.

Three captioning jobs

Before you publish

• Every spoken word that matters is on screen, and nothing was invented during music or crosstalk.
• Speaker names appear wherever a viewer couldn't tell who is talking.
• Sound cues describe sounds that matter, in one consistent style.
• No warnings remain in the line list, or you have checked each one on the preview.
• Captions stay clear of faces, graphics and platform buttons.
• The delivery matches the destination: an exported SRT or VTT for closed captions, a rendered MP4 where they must always show.

If you are captioning to meet an accessibility standard, the WCAG caption requirements explain what is expected. The editor helps you produce accurate captions; checking them against a standard is still your call.

About the subtitle editor on this page

Opens

MP4, MOV, M4V, WebM, MKV videos your browser can decode

Subtitles from

AI (from the speech, optionally translated into one of 21 languages), an SRT or VTT file, or typing them in

You download

An MP4 with the subtitles drawn into the picture at the original resolution and frame rate, plus SRT and VTT files

Cost

Editing, styling, rendering and SRT/VTT export are free, with no account and no watermark. AI subtitles cost 1 credit per minute (5-credit minimum per video); signing in with Google gives 100 free credits and accounts get 10 free credits a day

Privacy

The video never leaves your device and is rendered in your browser. Only for AI subtitles is the audio sent, over HTTPS, and it is deleted within 30 minutes of delivery

Rendering needs

Chrome or Edge 94+, Safari 16.4+, or Firefox 130+ on a computer. Editing and SRT/VTT export work in any current browser