PDF to Word
Convert text or scanned PDFs into editable Word documents. English and Korean OCR runs in your browser.
Your files are not uploaded to our server. All processing is done locally inside your web browser, and files are only saved to your device.
Drag and drop a PDF file here, or click to browse
Text or scanned PDF · Up to 100 MB · OCR up to 30 pages
⚠️Large PDF files may take longer to process depending on your browser's performance. We recommend keeping a backup of the original files before starting.
Practical guide
Convert text and scanned PDFs into editable Word text
Pages with selectable text use text extraction; pages with little or no text use OCR to produce editable paragraphs. English is the default OCR language; choose English + Korean for Korean documents. Use OCR every page when selectable text and pictures of text share a page.
The converter does not reconstruct table cells, column layouts, or image objects in Word. OCR is beta, and handwriting, equations, and blurry scans can be misread. The first engine and language-data downloads need an internet connection; document processing runs in your browser.
- Supported input
- Text PDFs and image-only scans
- Output
- DOCX (Word 2007 and later)
- Input limits
- 100 MB PDF · Up to 30 OCR pages
Good use cases
- ✓Editing document content you have permission to reuse
- ✓Rebuilding report paragraphs in Word
- ✓Recovering printed English and Korean text
Checks after download
- •Compare dates, amounts, and names against the source
- •Check reading order and missing content in tables and columns
- •Review original pages identified as having no recovered text
📝How to Use
- Choose a text PDF or scanned PDF under 100 MB.
- Choose English or English + Korean for OCR, then click Convert to Word. Pages without usable text are recognized automatically.
- For pages containing both selectable text and text inside pictures, select OCR every page. OCR supports up to 30 pages per conversion.
- The generated DOCX file downloads automatically to your device.
⚠️Caveats
- OCR (beta) recovers printed English and Korean text. Accuracy depends on scan quality; handwriting, equations and small text may not be recognized correctly.
- Tables, columns, images, and multi-column layouts are simplified into sequential paragraphs, so complex layouts need manual touch-up in Word.
- Complex layout conversion is experimental: table cells and original columns are not preserved. Fonts, colors and spacing may differ. Review the Word document before use.
- OCR downloads its engine and language data from jsDelivr on first use. Internet access is required for these downloads; your document stays in your browser.
❓Frequently Asked Questions (FAQ)
Q.Can scanned PDFs be converted to Word?
Yes. Pages with little or no selectable text use OCR to recover editable paragraphs. English is the default; choose English + Korean for Korean documents. Use OCR every page when a page mixes text and pictures of text. Tables, columns and images are not reproduced as their original objects.
Q.How does the converter rebuild paragraphs?
For text PDFs, items are grouped into lines by vertical position, then into paragraphs by line spacing. This can mix columns or move captions out of order. For scanned pages, paragraphs come from OCR. Compare the reading order with the source and edit it in Word when needed.
Q.Is the Word file generated on a server?
No. The DOCX package is assembled in your browser memory from the text extracted in your browser. The PDF is not uploaded, and the converted file is saved directly to your device.