PDF to Word Converter - With OCR for Scans

Turn a PDF into an editable Word document (.docx) with real paragraphs, headings and page breaks. Scanned pages without a text layer are read with OCR in seven languages. Everything runs in your browser: the PDF never leaves your device.

How a PDF Becomes a Word Document

A PDF stores text as strings placed at coordinates, not as paragraphs. This converter reads every string with its position and font size, groups strings on one baseline into lines, joins lines with a normal gap into paragraphs, and starts a new paragraph at a blank line, a bullet or a change of font size. Text set noticeably larger than the running text becomes a heading, so the result opens in Word with a real outline you can restyle, rather than one line per text box.

A scanned PDF has no text at all, only a picture of each page. When a page yields no text, the tool draws it at double resolution and hands it to Tesseract, the open source OCR engine, which runs as WebAssembly inside your browser. Pick the language of the document for the best result; the language model is downloaded once and cached. OCR on a clean, straight scan of printed text is reliable. Handwriting, skewed photos of pages and faint faxes are not, and the result then needs a read-through.

What is kept: paragraphs, headings, page order and page breaks. What is not: fonts, colours, columns, images and tables, which come out as running text. If you need the exact look of the PDF, keep the PDF. If you need to edit the words, this gives you a clean .docx to start from. The same conversion is available to scripts and AI agents through the ToolForte REST API and MCP server, where it takes a public URL and returns a download link, without OCR because there is no screen on a server to draw on.

How the PDF to Word Converter Works

  1. 01Select a PDF of up to 100 pages. Leave OCR on if it may contain scanned pages, and pick the language of the document.
  2. 02Click Convert to Word. Each page's text is read with its position and font size and rebuilt into lines, paragraphs and headings.
  3. 03Pages without a text layer are drawn at double resolution and read by Tesseract OCR inside your browser.
  4. 04Download the .docx. It opens in Word, Google Docs or LibreOffice with a real outline you can edit and restyle.

From Fixed Layout to Editable Text

A PDF places strings at coordinates and has no idea what a paragraph is. The converter groups strings on one baseline into lines, joins lines with a normal gap into paragraphs, and starts a new paragraph at a blank line, a bullet or a change of font size. Text that is clearly larger than the running text becomes a heading. A scanned page has no strings at all, so it is rendered to an image and passed to the OCR engine, which returns plain text that is split into paragraphs on blank lines. The result keeps the words, the reading order and the page breaks; it does not keep fonts, colours, columns, images or table borders.

When to Use This Tool

When someone sends you a contract, quote, report or form as a PDF and you need to change the wording, reuse a section or fix a typo. When you have a scanned letter or an old document and want the text back. When you want to paste a PDF's content into your own template without retyping. It is not the tool for preserving a designed layout: for that, keep the PDF and edit it with the PDF tools instead.

Common Use Cases

Tips

  • Switch OCR off for a PDF you know has a text layer: conversion is then instant.
  • For a mixed document (typed pages plus scanned attachments) leave OCR on; only the pages without text are scanned.
  • Very large scans take time: OCR runs at roughly one page every few seconds, all on your own device.

Frequently Asked Questions

Is my PDF uploaded?→
No. Reading the PDF, running OCR and writing the Word file all happen in your browser. Load the page, go offline, and it still works, except for the one-time download of the OCR language model. The REST API and MCP versions take a public URL instead, because they run on a server.
Why does the result look different from the PDF?→
Because a Word document flows and a PDF is fixed. The converter rebuilds paragraphs and headings so you can edit the text; it does not try to imitate the exact layout, which would give you dozens of text boxes that are worse to edit than a clean document.
How good is the OCR?→
On a straight, clean scan of printed text in the right language it is reliable. Photos of pages taken at an angle, faint faxes and handwriting give errors that need a read-through. Choose the document's language before converting: Dutch text read with the English model loses every accented letter.
Does it handle tables and images?→
Text inside tables is kept as running text in reading order; the table borders and images are not carried over. For a PDF that is mostly tables, the PDF text extractor or a spreadsheet import is often the better route.
Can I convert a password-protected PDF?→
No. Remove the password first with the PDF tool that made it, or ask the sender for an unlocked copy.

Related tools

12 suggested