OCR PDF

The tool runs Tesseract OCR server-side. It adds a searchable text layer over the existing scanned image without removing or replacing the original scan — the visual appearance of each page is unchanged. The text becomes selectable and Ctrl+F searchable. If you later want to extract the text to a plain .txt file, use the PDF to Text tool on the OCR’d output.

When OCR works well and when it doesn’t

Clean printed text on a flat, well-lit scan at 300 DPI or above: excellent accuracy. Handwriting: patchy at best. Mixed-orientation pages (some portrait, some landscape): the engine handles rotation before recognition, so this is usually fine. Very low contrast, coffee-stained, or heavily wrinkled documents: expect errors. The Enhanced mode applies pre-processing steps that help with marginal scans but won’t rescue a truly unreadable page.

Language support

Over 100 languages from a dropdown, including Arabic, Chinese (Simplified and Traditional), Japanese, Korean, Hindi, and all major European scripts. Select the language that matches the document’s primary script for best accuracy. Mixed-script documents — like a bilingual contract — can be processed but require the dominant language as the selection; the secondary script will get partial recognition.

OCR modes

Standard

Fast processing; reliable for clear, printed text. Most documents don’t need anything more.

Enhanced

Slower; applies additional pre-processing that improves recognition on low-resolution scans, faded ink, or pages with skew.