How to turn a scanned PDF into readable text
If a PDF is a photograph of a page, there may be no text to resize or search. Text recognition, also called OCR, gives the reader words to work with.
Check what kind of PDF you have
Try selecting a sentence in the source viewer. If the page behaves like a single picture, it may be image-only. Some files contain both a scan and an existing text layer; that layer can still contain mistakes.
Start with a readable scan
Use a straight, well-lit page with the whole text visible. Dark shadows, curved pages, cropped margins and low resolution make recognition harder. If the text is difficult for you to read, it is a poor starting point for automatic extraction.
Import and Restore
Save the scan as a PDF, import it into Ponte, then choose Restore. Ponte uses on-device text recognition when a usable text layer is missing. The Restore step is free and can work offline. It turns recognized text into a flowing reading view.
Proofread the result
Check the first paragraph, names, dates, numbers and unusual spellings. Watch for an “l” becoming a “1”, words split across line endings, and text from separate columns joining together. Handwriting, decorative type, equations and complex tables may need a different approach.
Translate only after checking
A recognition error can carry into a translation. If you need another language, check the extracted text first and start with a few pages. Cloud Translate requires internet access and paid credits; it is separate from the free Restore step.
Our sample pack includes a text PDF and an image-only version of the same page so you can compare the two workflows. They are labelled clearly and contain no personal information.
Free download, Restore and on-device reading. Cloud Translate and Explain use paid credits. iPhone and iPad, iOS 18 or later. Availability varies by storefront.