Will the DOCX look exactly like my original PDF?

No — this is a text-extraction-and-reflow tool, not a pixel-perfect layout clone. It pulls out the text, detects bold/italic and font size from each run, groups lines into paragraphs using their vertical spacing, and carries over embedded images as their own blocks in the same order they appear. It rebuilds all of that as an editable Word document with a page break between each original PDF page. Columns, exact positioning, and custom font files are not preserved — images are placed in reading order, not at their exact pixel position.

Why did I get an empty or near-empty document, or one with just images and no text?

That means the PDF (or that part of it) has no embedded text layer — most often because it's a scanned document made up of page images, not real text. There's no OCR (optical character recognition) in this tool, so it can only extract text that's already embedded in the PDF. For a scanned PDF, you'll still get the page images carried over into the .docx, but the text inside them won't be selectable or searchable — you'd need an OCR tool for that. A truly empty result means the PDF is empty, corrupted, or uses a feature this tool doesn't support.

How are paragraphs detected?

Text items on each page are stitched into lines using pdf.js's line-break signal, then consecutive lines are merged into the same paragraph unless the vertical gap between them is noticeably larger than a normal line's height — which usually means a blank line or extra spacing in the original document. It's a solid heuristic, not a perfect one, so occasionally a paragraph may be split or joined differently than you'd expect.

Is my PDF uploaded anywhere?

No. Text extraction runs with pdf.js and the .docx file is built with the docx library, both entirely in your browser — the file never leaves your device.

PDF TO DOCX

Extract a PDF's text and reflow it into an editable Word document. Nothing leaves your browser.

Drop a PDF here or click to browse

Nothing is uploaded — text is extracted right in your browser tab.

How to convert PDF to DOCX

1
Drop your PDF onto the upload area or click to select it.
2
The tool reads each page's text and images with pdf.js, groups text into paragraphs based on line spacing, and detects bold/italic and font size from each run.
3
Those paragraphs and images are assembled into a .docx file in their original order, one page break per original PDF page.
4
Click Download .docx to save the result.

Questions

Will the DOCX look exactly like my original PDF?: No — this is a text-extraction-and-reflow tool, not a pixel-perfect layout clone. It pulls out the text, detects bold/italic and font size from each run, groups lines into paragraphs using their vertical spacing, and carries over embedded images as their own blocks in the same order they appear. It rebuilds all of that as an editable Word document with a page break between each original PDF page. Columns, exact positioning, and custom font files are not preserved — images are placed in reading order, not at their exact pixel position.
Why did I get an empty or near-empty document, or one with just images and no text?: That means the PDF (or that part of it) has no embedded text layer — most often because it's a scanned document made up of page images, not real text. There's no OCR (optical character recognition) in this tool, so it can only extract text that's already embedded in the PDF. For a scanned PDF, you'll still get the page images carried over into the .docx, but the text inside them won't be selectable or searchable — you'd need an OCR tool for that. A truly empty result means the PDF is empty, corrupted, or uses a feature this tool doesn't support.
How are paragraphs detected?: Text items on each page are stitched into lines using pdf.js's line-break signal, then consecutive lines are merged into the same paragraph unless the vertical gap between them is noticeably larger than a normal line's height — which usually means a blank line or extra spacing in the original document. It's a solid heuristic, not a perfect one, so occasionally a paragraph may be split or joined differently than you'd expect.
Is my PDF uploaded anywhere?: No. Text extraction runs with pdf.js and the .docx file is built with the docx library, both entirely in your browser — the file never leaves your device.