Updated:
Copy words from a saved photo or screenshot with Live Text, or import the image into ScanLens to review the recognized text alongside the source. Check the copied result before reusing it.
Apple’s Live Text instructions cover copying from photos and enabling the feature. Availability depends on device, iOS version, language, and region.
Use ScanLens when you want the photo and reviewed OCR text together in your document library. This is a different storage and review workflow, not a guarantee of better recognition than Live Text.
Open the screenshot in Photos and use the same word-selection steps above. For tiny interface text, capture a new screenshot after enlarging the original content where possible. Enlarging an already blurry screenshot cannot restore missing detail.
No Live Text control: check Settings → General → Language & Region → Live Text. Apple lists iPhone XS/XR or later with iOS 15 or later as the minimum; supported languages and regions still vary.
Words will not select: try a clear printed line in another photo. Use the original image rather than a small message preview. Retake paper photos with even lighting, sharp focus, and the page facing the camera.
Text copies incorrectly: select a shorter passage and check column order, hyphenated lines, accents, and characters such as I/1 or O/0. OCR cannot reliably reconstruct letters missing from the image. See improving scan quality.
Paste extracted text into an editor if you want to rewrite it. A picture converted to PDF is still a picture unless text is added: use the searchable PDF guide when the text must travel inside the PDF.
For an image-only document, see photo to PDF or screenshot to PDF. For a broader OCR workflow and language coverage, see the OCR guide. Recognition in ScanLens runs on the iPhone; sharing and optional cloud backup use your selected destinations.
For another language, check the extracted words first, then translate the recognized document text.
Review both languages in the initial result. ScanLens offers a source-language override; if one script is missed, choose its language and recognize the image again. Check each passage rather than assuming that selecting one language guarantees both.
You can try recognition, but handwriting, connected letters, and mixed scripts may produce incomplete text. Check the output line by line; transcribe unclear words manually. The handwriting guide explains the limits.
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.
No sign-up, yes — ScanLens asks for no account and no email address at any point, and nothing here runs on someone else's server. Free, partly: recognition happens on every scan in the background and searching inside your own documents costs nothing, while the screen where you read, copy and export the recognized text belongs to the paid feature set. The pricing page has the current plans.