About
CapScribe turns a video or audio file into burned-in captions on iPhone or iPad. Transcription runs on Apple's on-device speech framework, and the app makes no network calls of its own β the only connection it opens is to the App Store for the purchase β so footage never leaves the device.
You get a word-timed transcript you can edit: find and replace, filler-word removal, and a review pass that flags low-confidence, overly fast or overly long segments. Edits keep the per-word timing. SRT and VTT import, and batches of up to ten clips.
26 caption presets across six styles, with per-word highlighting, keyword emphasis, stacked large-text layouts, custom fonts and an optional logo watermark. Placement is chosen per segment: it detects faces and on-screen text and puts captions where they won't cover anything. It does not identify anyone.
High-quality transcription covers about ten languages; other locales fall back to a basic-quality engine, which the interface labels as such.
One-time $8.99 unlock β no subscription, no credits, no watermark. The free tier covers the first 90 seconds of video or 10 minutes of audio. Requires iOS 26.