Photo to Text

Scan Text from a Photo on iPhone

Pull editable, copyable text out of any photo on your iPhone using on-device OCR via Apple's Vision framework in 50+ languages. Apple's built-in Live Text already covers quick captures from the camera; ScanLens earns its place when the language is outside Live Text's reliable set, when the photo is already sitting in your library, or when you want the recognized text saved alongside your other documents.

On-device Apple Vision 50+ languages Library integration No upload

Best for: photos already in your library, foreign-language signs and menus, pages of printed books, mixed-script documents (Japanese + English, Cyrillic + Latin), screenshots, and any photo whose recognized text you want stored next to related scans rather than just copied to the clipboard.

How do you scan text from a photo on iPhone — and is Live Text enough?

To scan text from a photo on iPhone, open the photo in an OCR-capable app, let the recognizer run, then copy, share, or save the extracted words. In recent iOS versions, Apple ships Live Text for exactly this — tap and hold on text inside any photo or the camera viewfinder, and a selection bubble appears with copy, translate, and share. For a single quick capture, that is the right tool.

A dedicated OCR app starts to matter when one of three things is true: the language sits outside Live Text's well-supported set (CJK accuracy, handwriting, mixed scripts), the photo has been sitting in your camera roll a while and you'd rather work from the library than the camera, or the recognized text needs to live in a document workflow — saved alongside other scans and PDFs, searchable across the library. ScanLens is the answer to those cases: on-device OCR via Apple Vision, 50+ languages, and a document library to keep the result.

Live Text vs ScanLens — when each wins

Live Text is the right tool when the text is in front of you and you want it in the clipboard now. A street sign you need to translate, a business card on the conference floor, a Wi-Fi password at a café, the model number on the back of a router. Open the camera, tap the text, copy. Done in three seconds. Live Text handles dozens of widely-spoken languages well, and translation is one tap away.

ScanLens earns its place in three distinct cases. A page of a book in Japanese you took two weeks ago — Live Text recognizes Japanese, but accuracy on dense vertical text is uneven and the result lives in the clipboard for thirty seconds before something else gets copied. A printed letter in your library — Live Text sometimes doesn't surface the selector on saved photos from months ago because the heuristics are conservative. A photo with two scripts side by side — a bilingual menu, a Russian-English instruction sheet, a Japanese-English textbook page. ScanLens's language picker lets you tell the engine what to expect, which is how dedicated OCR has always handled mixed documents.

What you get when you extract text from a photo in ScanLens

On-device OCR via Apple's Vision framework

The same engine Apple uses internally for Live Text, accessed directly with control over language selection. Vision recognizes 50+ languages including Cyrillic (Russian, Ukrainian), CJK (Chinese, Japanese, Korean), Arabic, Hebrew, Thai, Vietnamese, and handwriting in several Latin scripts. Nothing leaves the device — the photo and the recognized text both stay local.

Import any photo from library, camera roll, or Files

Pull a photo from the Photos app, import from a Files folder (iCloud Drive, Dropbox, Google Drive via the Files provider), or take a fresh shot inside ScanLens. Screenshots count — they're just photos as far as the OCR engine cares — which is useful when you need to grab text out of an app that won't let you select it directly.

Recognized text is editable, copyable, exportable

After OCR runs, the recognized text appears as an editable block. Copy the whole thing, copy a paragraph, share to Mail, Messages, Notes, or any iOS share-sheet target. Export as a plain .txt file if the destination prefers files to clipboard. Output is plain text — no formatting reproduction — which is what you usually want when extracting words from an image.

Saved alongside the photo in your ScanLens library

This is the part Live Text doesn't do. The recognized text is stored as a document inside ScanLens, linked to the source photo, and searchable from the library list next to your scans and PDFs. Two months later, when you remember you grabbed text off a Japanese restaurant menu but can't remember which photo, search finds it.

OCR Flow

How to scan text from a photo in ScanLens, step by step

From opening the app to having the recognized text on your clipboard is well under a minute for a single photo. Six steps, no roundtripping through other apps.

Step	Action	Tips
1	Open ScanLens	From the home screen or your iOS dock — no sign-in
2	Import the photo	Photos picker, Files browser, camera roll, or fresh capture inside the app
3	Pick the language(s)	For monolingual photos, default is fine; for mixed scripts, pick two or three
4	Wait for OCR (usually 1–2 seconds)	Recognized text appears as an editable overlay alongside the photo
5	Select, copy, or share	Long-press to select; standard iOS share sheet for Mail, Messages, Notes
6	Save the photo + text to your library	Stored as a single document, searchable next to your other scans

Photos where pulling out the text actually matters

Foreign-language signs and menus while travelling

A menu in a language you don't read, or a sign at a train station. Snap the photo, run OCR, paste into a translator. Live Text covers major European languages; ScanLens adds the long tail — Thai menus, Cyrillic timetables, Japanese hand-lettering — and keeps the photo and text together for after the trip.

Pages of a printed book or letter you photographed

A page in a library book you didn't have time to scan properly. A letter you photographed before mailing the original. A page of a friend's lecture notes. ScanLens turns the photo into text you can quote, search, or paste into a writing app.

Whiteboard photo from a meeting

The post-standup whiteboard with action items scrawled across it. OCR pulls the typed-and-handwritten mix into editable text so the action items can land in your task manager. For keeping the whiteboard itself as a clean visual document, see the whiteboard scanner flow with glare and perspective correction.

Recipe card from a cookbook

You don't want to scan-and-crop a whole cookbook for one recipe. Photograph the page, OCR the ingredients and method, save to your recipes folder. The text lives next to the photo so the visual layout is still there if a step is ambiguous.

Handwritten note you took a photo of

A sticky note someone left on your desk. A whiteboard sketch a colleague sent. A page of your own meeting notes. Apple Vision's handwriting recognition covers several Latin-script languages — for a dedicated walkthrough, see scan handwriting to text.

Frequently Asked Questions

Doesn't Live Text already do this?

For a quick capture — a street sign, a Wi-Fi password, a business card — Live Text is genuinely the right tool. It's built into stock iOS and works straight from the camera. ScanLens earns its place when the language is outside Live Text's reliable set, when the photo lives in your library, or when the recognized text needs to flow into a saved document rather than just the clipboard.

Which languages are supported?

ScanLens uses Apple's Vision framework for OCR, which recognizes text in 50+ languages — English, the major European languages, Cyrillic (Russian, Ukrainian), CJK (Chinese, Japanese, Korean), Arabic, Hebrew, Thai, Vietnamese, and handwriting in several Latin scripts. Multi-language documents are handled with a language picker so the recognizer knows which models to load.

Does the photo upload anywhere?

No. The OCR runs entirely on the iPhone via Apple's Vision framework. The photo never leaves the device for recognition. There is no ScanLens account and no upload step. If you save into an iCloud Drive folder afterwards, Apple's standard sync applies — your iCloud, not a ScanLens server.

What if the photo has multiple languages mixed?

Use the language picker to select the two or three languages that appear in the photo. This loads the right recognition models so the engine doesn't guess wrong on a Japanese-English menu or a Russian-English instruction sheet. Mixed-script accuracy is one of the main reasons people choose dedicated OCR over Live Text.

Can I scan text from a screenshot?

Yes — a screenshot is just a photo as far as ScanLens is concerned. Import from the camera roll like any other image. Useful for grabbing text out of an app that won't let you select directly, captioning a video frame, or pulling addresses out of a chat screenshot.

Will the recognized text reproduce formatting?

No. OCR output is plain text — words and line breaks, without bold, italics, font sizes, colors, columns, or tables. Text reflows in whatever app you paste into. If you need a styled copy, the workflow is OCR for the words plus a screenshot or PDF for the visual layout, kept side by side.