GekiPDF runs OCR (optical character recognition) over your scan and adds an invisible text layer, so the document becomes searchable and copyable while looking exactly the same.
OCR output is also the prerequisite for PDF to Word and PDF to Text on scanned documents. Chinese and English are both supported.
How to ocr pdf
- Upload the scanned PDF.
- Pick the language of the document.
- Press Run OCR (about 2–10 s per page).
- Download the searchable PDF.
Specifications
| Max file size | 50 MB |
|---|---|
| Languages | English, Chinese (Simplified), and both |
| Output | PDF with invisible text layer |
| Watermark | Never |
Frequently asked questions
Will OCR change how my PDF looks?
No. The text layer is invisible; the scan is untouched.
Which languages are supported?
English and Simplified Chinese work best. Choose “Chinese + English” for mixed documents.
How accurate is the OCR?
Clean 300 DPI scans typically reach 95–99% character accuracy; handwriting and low-resolution faxes are weaker.
Which free OCR handles Chinese scanned PDFs?
Tesseract with the chi_sim/chi_tra language packs, which is what GekiPDF uses. Pick eng+chi_sim for mixed documents; a single page of clean print usually takes about a second.
Does OCR change how the page looks?
No. OCR adds an invisible text layer behind the image, so the scan looks identical while becoming searchable and copyable.