Images & scanned documents

Recognize text with OCR

Turn scans into editable text and searchable PDFs. Review the result before using it.

12 languages · On your device

Choose images or one PDF

JPG, PNG, WebP or PDF · 10 MB per file

Up to 10 images or 10 PDF pages per run.

Turn a scan into text you can search and edit

OCR detects printed characters in a scanned PDF or image. Batchmint can export recognized text as Word or a searchable PDF, and lets you review the text before downloading.

How to use it

Choose one PDF or a batch of images, then select the document language.
Enter page numbers if you only need part of a PDF.
Run recognition and correct uncertain names, numbers and punctuation in the text preview.
Download a DOCX for editing or a searchable PDF for finding text later.

Limits and checks

The recognizer supports English, Arabic, French, German, Spanish, Portuguese, Italian, Russian, Hindi, Simplified Chinese, Japanese and Korean.
Each run accepts up to 10 images or 10 PDF pages; files must be 10 MB or smaller, with a 30 MB batch limit.
Clear, upright, high-contrast scans usually produce more reliable text than blurry or handwritten pages.
The document stays in your browser; the OCR engine and selected language data may need to load first.

Common questions

Does OCR reproduce the original page layout?

OCR extracts text; it does not recreate the original design. Keep the source scan for visual reference.

Can I search the exported PDF?

Yes. The searchable PDF keeps the page image and adds a text layer that search tools can read. Review the recognized text before relying on it.