Turn a photo, scan or screenshot into an editable Word document. The text is read on your device.
Click to upload or drop your images
JPG, PNG, WebP or a screenshot — add several and each becomes a pageOCR reads the words, not the layout — columns, tables and logos do not survive. Always read the text through before you send the document on.
About Image to Word
Point your phone at a page, or take a screenshot, and this tool reads the words and hands you a real .docx — one you can open in Word, Google Docs or LibreOffice and actually edit. Not a picture pasted into a document: proper, selectable, editable text.
The reading happens on your device with an OCR engine that runs in the browser. Your images are never uploaded, and there is no sign-up or watermark.
Add several pictures and each one becomes its own page, in the order you arranged them — handy for a multi-page document you photographed one page at a time.
What you can do with it
- Type up a printed page without typing it — photograph it and edit the result
- Get a screenshot into an editable document when you cannot copy the text
- Turn a scanned contract, letter or form into something you can amend
- Lift the text out of a book, sign or slide you photographed
- Read a page in another language into a document, ready to translate
What OCR can and cannot do
It reads words, not layout. That is the single most useful thing to understand before you start.
It handles well: clear printed text, straight-on photos, screenshots, dark text on a light background, and 12 languages including Chinese, Japanese, Korean, Russian and Vietnamese.
It struggles with: handwriting, columns (they run together), tables (they lose their grid), text over pictures, heavy skew, low light and blur. Logos and photos are ignored entirely.
That is why the text lands in an editable box before you download, not straight into a file. A minute of proofreading is normal and expected.
Features
- Real editable .docx — a proper Word file, not an image in a wrapper
- Several images become several pages, in the order you set
- 12 languages, including CJK, Cyrillic and Vietnamese
- Fix the text before you download — the result is editable on the page
- Optional headings — short capitalised lines become Word headings, so the document has structure
- Lists and addresses keep their shape — a wrapped sentence is joined back up, a deliberate line break is not. Tick the box to keep every break exactly as scanned
- Plain text download as well, if you only want the words
- 100% in your browser — nothing uploaded
Related tools
- Image to Text (OCR) — the same reading, but just the text to copy
- PDF to Word — start from a PDF instead of a picture
- Word to PDF — go the other way
- Image Compressor — shrink the photo before you scan it
How to use it
- Drop your images in — add several if you have several pages
- Choose the language the text is written in; this matters a lot for accuracy
- Press Read the text — the first run downloads the OCR engine, then it is cached
- Read through the result in the box and fix anything that came out wrong
- Press Download Word (.docx) — or take the plain text instead
