CapScribe

Captions burned into your video, entirely on your iPhone

CapScribe listens to your clip, writes timed captions, lets you fix any word, and renders the captions into the video file. Transcription, editing and rendering all run on your iPhone or iPad. Nothing is uploaded.

Download on the App Store Free preview, then a one-time purchase. No subscription, no watermark.

• Stays on your device. The app has no server. Your video, audio and captions never leave the phone.
• Buy once. One in-app purchase removes the preview limit for good. No credits, no monthly plan.
• Works offline. Airplane mode is fine once the language resources are on the device.

What you get

• 26 caption templates in six groups: classic, word by word, stacked big words, creator, minimal and neon. Four karaoke highlight styles: color, block, outline and scale.
• Keyword emphasis. Mark the words that matter and the template renders them bigger, bolder or in a second color. Every word can carry its own color.
• An editor that respects your edits. Fix a word and the timing of the other words stays put. Remove filler words in one tap. Undo, redo, find and replace, a review pane for low-confidence, too-fast and too-long lines.
• Smart placement. The app looks at each segment of the video and keeps captions away from faces and from text already on screen. Detection runs on the device and identifies nobody.
• Your fonts, your logo. Import TTF or OTF files, add a logo watermark, and save the whole look as a brand style.
• Translation and bilingual captions with Apple's system translation, plus SRT, VTT and TXT export when you want to finish in another editor.
• iPad included. Landscape split view with the preview beside the caption list.

How it compares

Most caption apps upload your video to a server, charge a monthly fee or a credit balance, and stamp a watermark on the free tier. CapScribe does none of that. The trade-off is honest: accuracy comes from Apple's on-device speech model, which is excellent in the languages it supports and basic in the rest. Editing before export is part of the workflow, and the editor is built for it.

Requirements

iOS 26 or later, on iPhone or iPad. The first time you transcribe or translate in a language, iOS may download the matching Apple language resources. Languages outside Apple's on-device model fall back to basic quality and are labeled that way in the app. The AI Assistant features from version 2.7 need a device with Apple Intelligence (iPhone 15 Pro or later, in supported regions).

What it costs

You can try the complete workflow at no charge. For longer media, the free preview processes the first 90 seconds of each video or the first 10 minutes of audio. A one-time, non-consumable in-app purchase removes the limit. The App Store shows the price for your region.

Support

Current limits

• iPhone runs in portrait. iPad supports both orientations.
• Accuracy depends on Apple's on-device speech model, which varies by language and by how clean the audio is.
• No speaker labels: Apple's framework does not provide them, and CapScribe does not send audio anywhere to get them.

---

字幕烧进视频，全程在你的 iPhone 上完成

CapScribe 听你的片子，写出带时间的字幕，让你改任何一个字，再把字幕渲染进视频文件。转写、编辑、渲染都在 iPhone 或 iPad 上跑，不上传。

在 App Store 下载 先免费试用，再一次性买断。没有订阅，没有水印。