Auto Captions in Seconds
Upload a video with spoken audio to create editable, word-timed captions. Accuracy varies with recording quality, language, and background noise.
Caption Settings
How to add auto captions to a video
1. Upload your clip
Drop in a talking-head video, podcast clip, or reel with spoken audio. Vertical, horizontal, or square — up to 240 MB.
2. Let the AI transcribe every word
Speech recognition timestamps each individual word, not just each line, so captions land exactly when they are spoken.
3. Edit, download, or burn them in
Fix any word right in the browser, download a standard
.srtsubtitle file, or take the video into the full NexusCut editor to style animated word-by-word captions and export a finished MP4 — no watermark on free exports.
Why captions matter for short-form video
A large share of social video is watched on mute — creator studies have long put it around 80%. Captions are how silent scrollers read your hook, and word-by-word animation keeps eyes on screen for the full clip. Auto captions turn a one-minute recording into an editable transcript in seconds, so the captioning step stops being the reason you do not publish.
Auto captions FAQ
Is the AI auto-caption generator free?
Yes. You can transcribe videos and download word-timed SRT files for free, with free usage limits that refresh every day. Paid plans add unlimited editing tools, priority processing, and caption burn-in exports.
How accurate are the auto captions?
Accuracy depends on the recording: clear speech in a quiet room transcribes very well, while background music, heavy accents, or overlapping speakers reduce accuracy. Every word stays editable, so you can fix anything the AI gets wrong.
Which languages does it support?
The transcription engine handles most major languages, including English, Spanish, French, German, Portuguese, Italian, and more. Set your expectations by testing a short clip of your own audio first.
What file formats can I upload and download?
Upload MP4, MOV, or WebM video up to 240 MB. Download your word-timed captions as a standard .srt subtitle file that editors like CapCut, Premiere, and YouTube accept.
How is this different from hard-coded caption apps?
Instead of only overlaying text, this tool gives you the raw word-level timing. You can download the SRT, or open the same video in the NexusCut editor to style animated word-by-word captions — fonts, colors, per-word highlight — and export a clean MP4 with no watermark.
Ready for captions people actually watch?