Add captions to TikTok videos
Drop a clip in above and the tool transcribes it on your device, syncs every word to the audio, and styles captions that read clean on a 9:16 phone screen. Edit any word, pick an animated style, and export a TikTok-ready MP4 or a subtitle file. It is free, runs entirely in your browser, and adds no watermark.
Captions built for the way TikTok is actually watched
TikTok is a vertical, full-screen, sound-optional feed. Much of the For You page is scrolled with the volume off or barely audible — on a commute, in a waiting room, in bed next to someone asleep. When the audio is muted, the on-screen text is the only thing carrying your message. On TikTok, captions are not a finishing touch; they are how the video communicates at all for most of the people who see it.
They also decide whether a viewer stays. The opening second is where TikTok wins or loses a watch, and a line of text that previews the hook gives a thumb a reason to stop scrolling. This tool is built around that: it puts readable, well-timed captions on your clip in a few minutes, so the message lands whether the sound is on or off.
AI transcription with word-level timing
Upload your clip and an on-device speech model detects the spoken language and transcribes it, producing timestamps for every individual word rather than rough blocks of text. That word-level timing is what makes captions feel native to TikTok — text that lands on the beat of fast talking, jump cuts, and punchlines instead of lagging a second behind.
The model handles roughly 99 languages. If it picks the wrong one, or you want captions in a different language than the audio, you can override the language and regenerate. Every word sits in an editable transcript, so you can fix a name, a slang term, or a brand spelling by typing over it, and the timing stays intact.
Your footage never leaves your device
This is a literal in-browser tool. The video you drop in is processed on your own machine. It is never uploaded, never sent to a server, and never used to train any model. The AI runs locally: it downloads once the first time you use it, then works offline on your hardware.
For TikTok that matters, because the things people caption are often unfinished, personal, or under embargo until they post — a draft, a face, a voice, an idea they do not want sitting on someone else's server. Here there is no server in the loop. Privacy is the architecture, not a line in a policy.
Four caption styles that suit the format
Pick from four animated styles, all tuned for vertical video. Karaoke highlights each word as it is spoken — the classic TikTok look that pulls the eye word by word. Highlighted drops the active word into a colored box for punchy, high-contrast emphasis. Minimal shows one clean word at a time with no distraction. Dynamic shows one word with a small pop, giving captions rhythm that matches quick edits.
Then make it yours. Choose a typeface (Inter, Montserrat, Oswald, Lora, or JetBrains Mono) and weight, set the size, and position the captions top, center, or bottom with a nudge to clear them of TikTok's username, caption, and right-side action buttons. Set text and highlight colors, add an outline and shadow so words stay legible over any background, and control how many words appear per line so nothing crowds the narrow 9:16 frame.
Export a TikTok-ready video or a subtitle file
When the captions look right, export them burned into the video as an MP4 at 1080 × 1920, the native 9:16 frame TikTok expects — so what you preview is exactly what uploads, with the styling baked in. Choose an output optimized for fast sharing or one that holds source quality, and the burn is hardware-accelerated so longer clips, up to TikTok's ten-minute ceiling, still process quickly. There is no watermark.
If you would rather caption inside TikTok's own editor or keep a transcript, download a .srt or .vtt subtitle file instead. Either way, fine-tune any line's timing before you export — slide a caption a few frames earlier or later until it sits on the audio.
Free, with no account and no paywall
Every style, every language, and every export is free for everyone. There is no sign-up, no account, no trial that expires, and no upgrade screen waiting at the export button. The MP4 you download is clean, with no watermark stamped across the corner of your video.
Free here is a deliberate choice about how the tool should work, not a stripped-down version of something better. You get the full transcription, the full set of styles, and full-quality export the first time and every time.
Questions
Not always. TikTok can add auto-captions and read a subtitle file, but burning them in gives you full control over the font, position, animation, and timing, and ensures they appear exactly as designed for everyone. This tool does both: export an MP4 with captions baked in, or download a .srt or .vtt to use in TikTok's editor.
Large enough to read at a glance on a phone, but not so large they get clipped by TikTok's interface. Keep captions clear of the bottom area where the username and caption sit and the right edge where the action buttons are — center or upper-center positions are safest. You can set the size, position, and per-line word count here and preview it in the real 9:16 frame before exporting.
No. Everything runs in your browser on your own device. Your video is never uploaded, never stored on a server, and never used to train any model. The AI model downloads once and then runs locally on your hardware.
It exports vertical 9:16 video at 1080 × 1920, which is TikTok's native frame. What you see in the preview is what you get in the file, so captions stay positioned correctly when you upload.
Yes. The tool auto-detects the spoken language across roughly 99 languages, and you can override it or regenerate the captions in a different language. Every word is editable afterward, so you can correct names, slang, and spelling without breaking the timing.