Drop photos, scans or screenshotsTap to choose photos, scans or screenshots
or click to browseor tap to choose
JPG, PNG, WebP or iPhone HEIC, one or many, in order. You can also paste a screenshot with Ctrl/Cmd+V. Read in your browser; nothing is uploaded.
Large files (over 50 MB) may take longer or run out of browser memory
Converts a photo, scan or screenshot of a document into an editable Word (.docx) file. The text is read with OCR in your browser and shown in an editable preview before you download, so you can correct misread words first. Add several pictures and each becomes a page, in the order you choose. Nothing is uploaded.
Drop JPG, PNG, WebP or iPhone HEIC pictures (or paste a screenshot with Ctrl/Cmd+V), put them in order, choose the language of the text and click Read the pictures. Edit the text in the preview (a blank line starts a new paragraph), choose US Letter or A4, and download the Word file. Tick "Straighten tilted photos" if the page was photographed at an angle.
The Word file contains the text as plain paragraphs: the original fonts, columns and tables are not recreated. If the picture is mostly a table, use Image to Excel instead. A flat, evenly lit, full-size photo reads best; handwriting is not supported; check numbers and codes by eye, because 0/O and 1/l are easy to confuse. The first reading downloads the language data once. A scanned PDF goes through PDF to Word.
- Will the Word file look like my photo?
- No. You get the text as ordinary paragraphs you can edit, not a copy of the layout. Fonts, columns, bold text and tables in the original are not recreated. If you want the picture in the document as well, tick the option to put the original picture above its text on each page.
- I have a phone photo of several pages. Can they become one document?
- Yes. Add the pictures in page order (drag to reorder); each becomes a page in the Word file. iPhone HEIC photos are accepted and turned upright automatically.
- The text has mistakes. Can I fix them before downloading?
- Yes, that is what the preview is for. Each picture's text sits in a box you can edit, and a blank line starts a new paragraph. The Word file is built from what is in the boxes. Numbers and codes need a careful look, because 0 and O, or 1 and l, are easy to confuse.
- What if the photo is tilted or sideways?
- A photo that is sideways or upside down is turned upright automatically when the first reading is poor. For a page photographed at a slight angle, tick Straighten tilted photos before reading. Leave it off for screenshots.
How to use Image to Word
Add one picture or several of a document: JPG, PNG, WebP or an iPhone HEIC photo, or paste a screenshot. Put them in page order and choose the language. Press Read the pictures, then edit each picture's text in the preview (a blank line starts a new paragraph). Choose US Letter or A4, decide whether the original picture goes above its text, and download the .docx.
When to use it
Use it for a photographed letter or contract page, a screenshot of an article or a chat, a page of a book you need to quote, or meeting notes written on a whiteboard in clear print, whenever you need the words in Word to edit.
When not to use it
If the picture is mostly a table, use Image to Excel so the columns survive. If the file is a PDF, use PDF to Word, which keeps real fonts, lists and tables when the PDF has a text layer. For handwriting, no setting here will help.
Tips for best results
Tick Straighten tilted photos for a page photographed at an angle, not for screenshots. Fix 0/O and 1/l mix-ups in numbers in the preview before you download. For a long document, photograph the pages in order and check the thumbnails before you read.
Limitations
The Word file is plain paragraphs: no original fonts, columns, bold or tables. Pages with no readable text are kept as a picture instead. Up to 30 pictures, 25 MB each. Reading takes several seconds a picture, and the first reading downloads the language data.
Privacy
Your pictures are read and the file is built entirely in your browser; nothing is uploaded or stored. The one network request is the first-time download of the text-recognition language data, which your browser then keeps.
Related document tools: