CapiCap: Video Captions

Whisper transcribes your video right on your iPhone or iPad, then burns clean subtitles in. No watermarks, no accounts — and nothing you make ever leaves your device.

CapiCap adds subtitles to your videos — automatically, and 100% on your device.

Free forever. No watermarks. No subscriptions. No account. No tracking. Nothing you make ever leaves your device.

Pick a video and CapiCap transcribes it locally using Whisper (the same open speech model that powers professional tools), running entirely on your device's Neural Engine. There is no upload, no server, and no waiting in a queue — no network connection required at all.

HOW IT WORKS

1. Pick a video from your library.
2. CapiCap transcribes it on-device, faster than real time on a recent iPhone or iPad.
3. Fix any words in the built-in editor.
4. Style your captions — fonts, colors, size, position, and pill or band backgrounds.
5. Export a new video with the captions burned in, ready to share.

Prefer a subtitle file? CapiCap also exports a standard .srt you can use anywhere.

WHY CAPICAP

• Private by design. Transcription happens entirely on your device — your videos and words are never uploaded.
• Works offline. No signal, no problem. Everything runs locally.
• iPhone and iPad, one app. On iPad the editor gets room to breathe: a big preview with the whole caption list beside it.
• No watermarks, ever. Your exports are clean.
• Genuinely free. No paywalls, no "pro" upsell, no ads.
• Simple styling that looks good straight out of the box.

A LITTLE SUPPORT CapiCap is free and always will be. If it saved you some time, there is an optional tip jar with a few capybara-themed tiers — a small thank-you that unlocks nothing. Tips are pure gratitude.

REQUIREMENTS Requires iOS 17 or later, on iPhone and iPad. CapiCap matches the speech model to your device: iPhone 13 and later, and iPad Air 5 and later, use Whisper Large v3, the most accurate model; older devices use the smaller Whisper Base. It downloads once on first launch. Works best on iPhone 14 Pro or newer and on iPad with an M-series chip, which transcribe faster than real time.