Recognize text in a scanned PDF

A scanned PDF often contains page images rather than selectable words. Optical character recognition (OCR) reads those images and creates text you can edit or search. Recognition is an estimate: always compare important names, numbers and dates with the page image.

Prepare a clear source

• Use an upright, sharp scan with good contrast.
• Keep text large enough to read and avoid heavy compression.
• Use one PDF or a batch of images, not mixed PDFs and images.
• The tool accepts PDF, JPG, PNG and WebP, up to 10 MB per file and 30 MB per image batch.

Select only the pages you need

For a PDF, enter page numbers or ranges such as 1-3, 5. You can recognize up to 10 pages in one run. For image batches, choose up to 10 images. Smaller selections take less time and make review easier.

Choose the language and check the text

Batchmint includes English, Arabic, French, German, Spanish, Portuguese, Italian, Russian, Hindi, simplified Chinese, Japanese and Korean recognition. Select the main language; for a document that also contains English, you can enable English alongside another language. Mixed scripts, handwriting, tables and low-resolution text may need manual correction.

Choose an output for your next step

Download editable text as Word when you need to revise the wording. Choose a searchable PDF when you want to keep the page images with a searchable OCR text layer. Review both outputs: editing recognized text does not rewrite the original image on the scanned page.

The first run downloads the recognition engine and language data to your device. The file itself is processed in the browser and is not uploaded by Batchmint.