OCR text extraction from images and PDFs

A free browser-based OCR tool for extracting text from images and PDFs. Runs the full Tesseract engine locally via WebAssembly, supports more than 100 languages, fixes sideways photos and shadowed scans automatically, and exports to TXT, Word, Excel or a searchable PDF with an invisible text layer.

How to extract text from an image or PDF

1
Add files

Drop images or PDFs, shoot with the camera on mobile, paste from clipboard with Ctrl+V, or paste an image URL.

2
Pick the language

The default covers the site language plus English; choose up to three languages from the grouped dropdown for other documents.

3
Choose output style

Plain text for copy-paste, Preserve layout for documents, Extract tables for spreadsheets, or Receipt for totals, dates and vendors.

4
Review and edit

Inspect the bounding boxes, fix any low-confidence words inline, search for specific words.

5
Export or copy

Download as TXT, DOCX, XLSX or searchable PDF — or bundle a whole batch into a ZIP.

Turn scans, photos and PDFs into editable text, Word, Excel or searchable PDF

Drop images or PDFs here or browse

Or press Ctrl+V to paste a screenshot · type / paste URL

Working…

High confidence Medium Low — review

Click a box to jump to the word · click the word to highlight the box · Edit any word inline — your corrections are saved to the export.

QR code detected:

Export

Runs in your browser via WebAssembly. Works offline after the first use.

Keyboard shortcuts

Ctrl+VPaste an image from the clipboard Ctrl+FFind in extracted text Ctrl+EnterStart extraction EscCancel current operation [ / ]Jump to next low-confidence word Ctrl+scrollZoom the image preview
Frame the document inside the guide · hold steady · tap to capture

What makes this OCR different

100+ languages
English, Russian, Chinese, Japanese, Korean, Arabic, Hindi and 100+ more — pick one or mix up to three for bilingual documents.
Batch up to 50 files
Process up to 50 images and multi-page PDFs per batch on a parallel worker pool; large PDFs get a page-range picker.
Four export formats
Plain TXT, Word (.docx) with formatting, Excel (.xlsx) with detected tables, or a searchable PDF with an invisible text layer.
Smart preprocessing
Sideways photos are auto-rotated, shadowed or washed-out shots get adaptive binarization, skewed scans are straightened — all automatic.
Bounding-box editor
See which word came from where, low-confidence words highlighted in yellow, click a box to correct inline before export.
Camera & paste input
Shoot with your phone camera, paste screenshots from the clipboard, or fetch an image from a URL.

What this tool is good for

  • Students — digitise lecture notes and textbook scans into searchable text
  • Accountants and bookkeepers — pull line items from receipts and invoices into Excel
  • Archivists — convert decades of paper scans into a searchable PDF library
  • Researchers — extract quotes from screenshots and foreign-language sources
  • Translators — copy source text out of an image without retyping
  • Developers — grab code or logs from screenshots, lift text out of UI mockups

Supported input and output

Input: JPG, PNG, WebP, TIFF, BMP, HEIC, PDF (multi-page). Output: TXT, DOCX, XLSX, searchable PDF, Markdown, JSON with bounding boxes.

Frequently asked questions

How accurate is the OCR?

Clean printed text recognises nearly perfectly; phone photos and faded scans are harder. The automatic pipeline straightens skew, fixes sideways shots and applies adaptive binarization to shadowed photos, and any low-confidence words are highlighted so you can correct them before export.

Which languages are supported?

All 100+ languages from the Tesseract 4 tessdata set: English, Spanish, French, German, Russian, Chinese (Simplified & Traditional), Japanese, Korean, Arabic, Hindi, Bengali, Thai, Vietnamese, Turkish, Polish and dozens more. Languages are grouped by script in the dropdown; models load on first use and cache in your browser.

Does it work with PDFs?

Yes. Multi-page PDFs are rendered page by page via pdf.js and then passed through OCR. You can also export the result back as a searchable PDF with an invisible text layer, so Ctrl+F works and screen readers can read it.

Can it recognise handwriting?

Partially. The underlying engine is tuned for printed text; neat handwriting with clear letter separation is often readable, cursive is not. If handwriting accuracy is critical, we recommend a dedicated handwriting OCR service.

How does table extraction work?

After OCR we group recognised words by their vertical position into rows and split rows into columns based on horizontal gaps. When at least three rows share the same column count we treat it as a table and export it as rows and columns in Excel; otherwise the Excel file has a single column of text.

Are there any file size or page limits?

Up to 50 files per batch; the per-file size cap adapts to your device (roughly 80 MB on phones up to 1 GB on desktops). PDFs stream page by page, so long documents stay memory-safe, and past 150 pages a range picker lets you OCR exactly the pages you need.

How do you rate this tool?

Thank you for your rating!
Want to share more? Leave a comment!
Thank you! Your comment will appear after moderation.
Who is this tool for?
✨ Content Creators 🎓 Students 📚 Educators & Teachers 💻 Developers & IT Pros 📷 Photographers
Published Updated Authors: