Opening tool…
Keep Lyonite open while connected to make the site available offline.
Opening tool…
Convert text, images and English scans into an editable Word document. Everything happens on this device.
Runs on your device. Nothing is uploaded.
Drop PDFs here
PDF or ZIP · up to 200 MB · processed on this device
Select or drag it here. It is not uploaded
Text, columns, tables, images and scans are read page by page
Take it as .docx, as .txt, or copy it
Download gives you a .docx file, which is the format Word has used since 2007 and which Google Docs, LibreOffice and Pages all open directly. If you specifically need the older .doc format, open the .docx in Word and save it as .doc. Check the layout after saving.
The same extraction backs the .txt download and the copy button, so if all you want is the words — to paste somewhere, to search, to feed into something else — take the text and skip the Word file entirely. It comes out in reading order, page by page, with the line breaks the PDF actually contains.
A scanned PDF is a picture of a page, so there is no text to copy until something reads it. When a page has no text layer, Lyonite reads it with English OCR in your browser and puts the recognized words into the Word file as ordinary, editable text. Uneven lighting from a phone photo is evened out before reading, and two-column scans are read one column at a time.
On three test scans — a clean 300 dpi page, a skewed and noisy two-column page, and a typewritten 1967 report — it recovered 94 to 97 percent of the words. On the same day the free PDF-to-Word converters of iLovePDF and PDF24 returned all three as pictures, and Smallpdf’s OCR is a paid feature. See the full comparison.
Proofread the result. Handwriting, very small print and languages other than English are not read reliably, and anything the reader is unsure of is flagged under Conversion notes.
Yes. The same extraction backs the .txt download and the copy button, so you can take the words on their own without producing a Word file at all.
It produces .docx, the format Word has used since 2007, which Google Docs, LibreOffice and Pages open directly. For the older .doc format, open the .docx in Word once and save it as .doc.
Documents such as reports and papers become editable paragraphs, with columns read in order and tables rebuilt where they are detected. Forms, slides and ruled tables keep their layout: each line is placed where it was, over the page’s lines and shading. Complex title pages and scanned text should be reviewed.
Yes, for free. English OCR automatically reads scanned pages and scanned regions on your device. On three test scans it recovered 94 to 97 percent of the words. Review recognized text, especially on blurry or photographed pages. OCR works offline after the tools finish preparing.
Yes. Choose the PDF above and download the .docx. The text is editable in Word, Google Docs, LibreOffice and Pages, including text read from scanned pages. Filled-in form values are kept.
Your PDF never leaves your browser. The entire conversion process runs locally using JavaScript. No data is sent to any server, ever. Once you close the tab, the file is gone from memory.