Turning scans into text with OCR in the browser

What OCR actually does

OCR looks at the shapes on a page and guesses which characters they are. It is a recognition process, not a conversion, so the output is always an estimate — clean typed pages come out near-perfect, while faint faxes and handwriting do not.

PDFzy runs recognition with Tesseract compiled for the browser. The language model downloads once and is cached by your browser; your document itself is never sent anywhere.

Getting better accuracy

Small changes to the input make a large difference to the result.

• Scan or photograph at 300 DPI where you can.
• Straighten the page before recognition; tilted lines confuse the engine.
• Pick the correct language — accented characters depend on it.
• Increase contrast for grey or shadowed photos.

The current limitation

PDFzy gives you the recognised text to copy, save, or paste. It does not yet write an invisible text layer back into the original PDF, so the tool produces text rather than a searchable PDF.