Free browser tool

PDF to Word

Extract selectable text page by page, correct it in an accessible editor, and create a real Word document or plain-text download.

Processed on your device

PDF.js reads this file in a module worker. The document itself is not uploaded to EditableDocument, Cloudflare, an advertising provider, or an analytics service.

Choose a PDF to extractOne text-based PDF, up to 50 MB

Simple workflow

How it works

Select one PDF

Choose or drop a text-based PDF up to the 50 MB browser reliability limit.

Extract the text

A worker reads each page and reports progress without blocking the main page interface.

Review and edit

Correct line breaks, reading order, headings, and other details in the editor.

Download your result

Create a DOCX file for continued editing or a simple TXT file.

Compatibility

Supported formats

Input: PDF

Single application/pdf file with a valid PDF header and a selectable text layer, up to 50 MB.

Output: DOCX or TXT

DOCX contains editable paragraphs; TXT preserves the corrected plain text without styling.

Honest expectations

Known limitations

PDF is a fixed-layout format, so positioned fragments must be reconstructed into reading order.
Complex columns can be interleaved, and tables can lose their row-and-column structure.
Images, forms, annotations, custom fonts, and precise page geometry are not reproduced.
Scanned pages require OCR, which this tool does not include.
Password-protected PDFs are not supported.
Large files may exceed the memory available to a mobile browser.

Document privacy

What stays in your browser

The worker receives an in-memory copy of the file selected through your browser. It extracts text locally and terminates after completion, cancellation, or an error.

If you edit the result, a lightweight draft can be saved in IndexedDB so you can restore it on this device. That draft contains editable content, not the original PDF binary. Use the discard control to delete it.

Questions answered

Frequently asked questions

Is the PDF uploaded for conversion?

No. PDF.js reads the selected file inside a browser worker. The extracted text is returned to the editor on this page; the PDF is not transmitted to EditableDocument.

Why did the columns appear in the wrong order?

A PDF stores positioned text fragments rather than an editable reading structure. The extractor uses page coordinates to rebuild lines, but multi-column layouts can still be interpreted in an unexpected order.

Can this tool convert a scanned PDF?

No. A scanned page is usually an image without a selectable text layer. OCR is not included, so the tool reports when no extractable text is found.

Can I edit the extracted text?

Yes. Use the built-in editor for headings, paragraphs, bold, italic, lists, undo, and redo before downloading DOCX or TXT.

Where is a restored draft stored?

Lightweight editor content is saved in IndexedDB on this device after you edit. The uploaded PDF binary is not stored. You can discard the draft from the tool.

Keep working

Practical reading