How OCR works in ConvertiZen
OCR analyzes the pixels in your image and recognizes the characters. ConvertiZen uses Tesseract.js — the leading open-source OCR engine, compiled to WebAssembly — so recognition happens locally on your device at high speed without any server-side processing.
FAQ
What languages does the OCR support?
English, French, Spanish, German, and 50+ other languages.
Does OCR work on low-quality scans?
Best results with clear, high-contrast scans. Poor quality reduces accuracy.
Are my images sent to a server for OCR?
No. Tesseract.js runs locally via WebAssembly. Images stay on your device.
Can I run OCR on a scanned PDF?
Yes, ConvertiZen can extract text from scanned PDF pages.
Is table structure recognized?
Text is extracted accurately; table structure may be partially lost.