How do I add captions to a video for free?

1. Upload an MP4, MOV, or WebM up to 500 MB and 10 minutes long. 2. Choose a caption style, or leave it on Auto. 3. We transcribe the speech, time every word, and burn the captions into the video. 4. Download the captioned MP4. There's no software to install, no account to make, and no watermark on the file you get back.

Is it really free, and is there a watermark?

It's free, and there's no watermark. The MP4 you download is the finished video, not a preview. You don't need an account, an email address, or a card, and there's no export you have to upgrade to unlock. Most caption tools give you a free tier that stamps the export or caps it well below a usable length; this doesn't.

What's the difference between captions and subtitles?

Captions are the same language as the speech and are written for viewers who can't hear the audio, so they cover what's being said in the video's own language. Subtitles usually mean a translation into another language. This tool makes captions: it transcribes what was said and keeps it in the original language. If you want a video translated into a different language, use our AI video translator instead.

How accurate are the captions?

Captions are a straight transcript of what was said, produced by Azure AI Speech. There's no language model in the pipeline, so nothing gets paraphrased or invented. Accuracy depends on your audio: clear speech with little background noise transcribes well, while heavy accents, crosstalk, loud music, or a distant microphone all make it harder. We'd rather point you at what affects the result than quote an accuracy figure we haven't measured on your footage.

Can I edit the captions before I download?

Not at the moment. The tool goes straight from upload to a captioned MP4, which is what makes it a three-click job rather than a session in an editor. If a word comes out wrong, you'd need to re-record or use a full editor. Caption editing is something we're looking at.

Can I download an SRT or VTT file?

No. The captions are burned into the video, so they're part of the picture and play the same way everywhere, on any player and on any social platform, without you uploading a separate file. That also means there's no sidecar subtitle file to download and no way to switch the captions off. If you specifically need an SRT, our AI video translator gives you subtitle files alongside the video.

Which languages are supported?

Speech is recognised in ten languages: English, Spanish, French, German, Italian, Brazilian Portuguese, Hindi, Japanese, Mandarin Chinese, and Arabic. The language is detected from the audio automatically, so there's nothing to select. Recognition is tuned to those specific variants, so a language outside that list won't transcribe well. Captions come back in whatever language was spoken; the tool doesn't translate between languages.

What video formats and sizes can I upload?

MP4, MOV, and WebM, up to 500 MB and 10 minutes long. The video needs an audio track with speech in it, and we check for one before starting, so you find out straight away instead of waiting on a job that can't work. Whatever you upload, you get an MP4 back, encoded as H.264 with AAC audio. Anything larger than roughly 1080p is scaled down to keep the file manageable.

Will the captions work on TikTok, Reels, and Shorts?

Yes. Because the captions are burned into the picture rather than attached as a separate file, they survive every upload and re-encode and show up wherever the video plays. Vertical and widescreen video are both handled, with caption size and line length set to suit the shape of the frame, and captions placed clear of the bottom edge where platform interface elements sit.

How long does captioning take?

It depends on how long your video is, because adding captions means re-encoding the whole thing rather than attaching a file to it. A short social clip comes back much sooner than a ten-minute one. The page keeps you posted as it works through transcribing, preparing the captions, and rendering.

Free Video Caption Generator - No Watermark, No Sign-Up

How it works

Add captions to a video in three steps

No editor, no timeline, no account. Upload a video and download it captioned.

Upload your video

Drop in an MP4, MOV, or WebM up to 500 MB and 10 minutes long. The video needs an audio track with speech in it.

We transcribe and time every word

Your speech is transcribed with word-level timings, then split into captions of usually six to twelve words. The spoken language is detected for you.

Download the captioned MP4

Captions are burned into the video, so they play everywhere without a separate file. You get an MP4 back with no watermark.

What you get

Captions that keep pace with the speaker

Most caption tools drop static text on screen. These light up word by word, in time with the voice.

Every word lights up as it's spoken

Each caption starts dim, and every word fades to its full colour at the moment it's said. The timing comes from word-level speech timings, not an even split across the line, so it tracks pauses and fast speech alike. When the last word lands, the next caption takes over.

Presets, or your own two colours

Choose captions on a rounded background or outlined straight onto the footage. Both modes ship colour presets, and custom lets you set the text and backdrop yourself. The dim colour is worked out from those two, so there's never a third picker to fiddle with.

Colours that adapt to your footage

On Auto, the brightness behind the caption is measured for every caption and the colours are chosen to suit it. A clip that cuts from a bright room to a dark one gets readable captions in both, without you touching a setting.

Free, with nothing held back

No watermark on the export, no account, no email, and no card. The captioned MP4 you download is the finished file, not a preview you have to pay to unlock.

Why SlideSpeak

The details that make captions readable

Line length, timing, contrast, and placement all affect whether a viewer can read your captions. Most tools leave them to chance.

A faithful transcript

There's no language model rewriting your words. Captions are what the speaker said, so nothing gets paraphrased, tidied up, or invented.

The language is detected for you

Speech is recognised across ten languages: English, Spanish, French, German, Italian, Brazilian Portuguese, Hindi, Japanese, Mandarin Chinese, and Arabic. There's no dropdown to get wrong.

Lines sized for reading at a glance

Captions wrap to a column about 25 characters wide on vertical video and 37 on widescreen, measured from the real width of each letter rather than counted, so a line never runs past what you can take in.

Timing that respects reading speed

Captions are held toward a 0.9-second floor so they don't flash past, and the gap between two of them is either closed outright or left long enough to read as deliberate.

An outline that keeps text legible

With the background off, captions carry an outline against the footage, so even when the shot is busy or the brightness guess is imperfect, the text stays readable instead of disappearing.

Placed clear of the frame edge

Captions sit inside the caption safe area at the bottom of the frame, clear of descenders being clipped and of the lower-third graphics many clips already carry.

The specifics

The limits, published up front

Most caption tools don't tell you what they'll accept until your upload fails. Ours are below.

Languages, detected automatically

500 MB

Max upload, up to 10 minutes

Watermarks

Free

No sign-up needed

Comparison

How it compares

Every tool here calls itself free. The difference is what the free export gives you. Checked against each vendor's own pricing page in July 2026.

Features		VEED	Kapwing	Submagic
No watermark on free exports	Never	Watermarked	Watermarked	Watermarked
No sign-up or account	Not needed	Required	Required	Required
Word-by-word animated captions	Included	Included	Included	Included
Edit captions before export	Not yet	Full editor	Full editor	Text editor