How do I convert an image to a Word document?
Add the image above. Its text is read in your browser and shown next to the image as a Word preview. Check it, correct anything that's wrong, and select Download Word to save an editable .docx file.
Free OCR in your browser
Image to Word Converter
Turn photos, scans and screenshots into an editable Word document, with headings and paragraphs where they were in the image.
Last updated: October 2026
.docx
Editable Word file
Many
Images in one file
29
Languages
Private
No uploads
Drop your images here
JPG, PNG, WebP, HEIC, TIFF, BMP, GIF or AVIF. Several images make one Word file.
Your files never leave this device
or paste a screenshot with Ctrl+V
How to convert an image to Word
01
Add your images
Drop, choose or paste JPG, PNG, WebP, HEIC, TIFF or other images, or take a photo on your phone. Each image becomes a page, and a multi-page TIFF adds every page.
02
Check the language
Pick the language written in the image, and a second one if the text is mixed. Reading starts by itself, in your browser.
03
Review and correct
Compare each page with its Word preview. Tap a paragraph to find it on the image, and fix any word before you download.
04
Download the Word file
Choose Letter or A4 and download one .docx with every page, ready to edit in Word.
What goes into the Word file
The text is read with optical character recognition (OCR), which turns the letters in an image into real text. Plain OCR gives you one long run of lines. This converter also works out how the page is built: which lines are headings, where paragraphs start and end, what is centered or indented, and which column comes first. The result is a .docx file in the Office Open XML format that Word uses.
Real headings
Larger text becomes Word's Heading 1 and Heading 2, so it shows in Word's navigation pane.
Paragraphs, not broken lines
Lines join into paragraphs, and a word hyphenated across two lines is joined back up. Or keep each line as it is.
Alignment and indents
Centered titles, right-aligned dates, first-line indents and hanging indents on lists are kept.
Several images, one file
Each image is a page. A multi-page TIFF adds every page, in order.
Right-to-left text
Arabic, Hebrew, Persian and Urdu paragraphs are set right to left in Word.
Words to check
Words the engine wasn't sure about can be highlighted in yellow, in the preview and in Word.
Your corrections
Fix a word or delete a paragraph in the preview, and the file has your version.
The original image, if you like
Each image can follow its text on a page of its own, so you can compare in Word.
What isn't converted
It helps to know the limits before you start, so nothing in the Word file is a surprise.
Bold, italic and fonts
The reading engine reports the text, where it is and how big it is, not its style. Headings are found by size, and all text uses Word's standard font.
Tables
A table's rows come out as lines of text, not a Word table. Check the order of columns.
Pictures and logos
Only text is converted. Turn on Add the original image to keep the whole page as a picture too.
Handwriting
Printed text reads well; joined handwriting usually doesn't.
How to get the most accurate Word file
Photograph the page straight on
Hold the phone parallel to the page, in even light, and fill the frame with the text. Shadows and steep angles cost more accuracy than anything else.
Turn it upright
Use the turn buttons for sideways pages. Tesseract's documentation notes that its line finding reduces significantly when a page is too skewed.
Use enough resolution
For scans, 300 DPI is a good choice: Tesseract's documentation says it works best on images of at least 300 DPI. Avoid screenshots of screenshots and photos sent through chat apps, which are compressed.
Crop to the text
A tight crop of the part you need removes clutter that can be read as stray letters.
Clean up a dark or gray photo
The image enhancer's Text or document mode evens out shadows and whitens the paper before you convert.
Pick the right language
The language decides which letters the engine expects. The wrong one gives nonsense even from a perfect scan.
To prepare a photo, use Crop Image to cut out the text you need, or the image enhancer's document mode to even out shadows. Sources: Tesseract documentation, Improving the quality of the output.
Languages, including right-to-left text
The converter reads English, Arabic, Bengali, Chinese (Simplified and Traditional), Dutch, French, German, Gujarati, Hebrew, Hindi, Indonesian, Italian, Japanese, Korean, Marathi, Persian, Polish, Portuguese, Punjabi, Russian, Spanish, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu and Vietnamese. Choose the language written in the image. If the page mixes two, such as Hindi with English, add the second one, and the engine looks for both. The text isn't translated.
Arabic, Hebrew, Persian and Urdu paragraphs are set right to left in the Word file, and the document's language is set so Word's spelling check uses the right dictionary. Each language downloads once, the first time you use it; English is about 3 MB.
This converter or Google Docs?
Google Drive can also turn an image into text: right-click the file and choose Open with, then Google Docs. Here is how the two compare, with Google's figures from its own help page. Microsoft Lens, which could save a photo as a Word file, was retired from Android starting January 9, 2026, and Microsoft stopped supporting it after February 9, 2026.
| Feature | This converter | Google Docs |
|---|---|---|
| Where it runs | In your browser; nothing is uploaded | Upload to Google Drive (Google account) |
| File size | No fixed limit (large images are read at up to 12 megapixels) | 2 MB or smaller |
| Images | JPG, PNG, WebP, HEIC, TIFF (every page), BMP, GIF, AVIF; many at once | JPEG, PNG, GIF and PDF |
| Kept | Headings, paragraphs, alignment, indents, right-to-left text | Bold, italics, font size, font type and line breaks are likely kept |
| Not kept | Bold, italic, fonts, tables | Lists, tables, columns, footnotes and endnotes are unlikely to be detected |
| Before saving | Preview beside the image, corrections, highlighted words | Opens as a Google Doc to edit |
Sources: Google Drive Help, Convert PDF and photo files to text; Microsoft Support, Microsoft Lens for Android.
Private by design
The text is read by Tesseract.js, an open-source OCR engine, running in your browser. The engine comes from ImgTweak and the language data from the jsDelivr network the first time you use them; after that your browser can keep them. Your images, the text and the Word file stay on your device and are never uploaded.
Pages are read one at a time, so a phone doesn't have to hold several large photos at once. The first page takes longest while the engine starts; the rest follow faster. Need just the plain text of one image? Use Image to Text. Have a scanned PDF? Turn its pages into images with PDF to JPG first.
Common image-to-Word mistakes
Choosing the wrong language
Hindi read as English, or Arabic read as French, comes out as nonsense. Pick the language in the image, and add a second one for mixed text.
Using a compressed copy
Images sent through chat apps or screenshotted again lose the fine edges of letters. Use the original photo or scan.
Expecting an exact copy of the page
The Word file is editable text with its structure, not a picture of the page. For an exact copy, make a PDF with Images to PDF instead.
Skipping the check
OCR can confuse O and 0, l and 1, or rn and m. Check names, numbers, dates and amounts against the image before you use the text.
Why use ImgTweak?
✓No upload, no account
✓Several images in one .docx
✓Headings and paragraphs kept
✓29 languages, right-to-left included
✓Correct the text before you download
✓Works on phones, with the camera
Frequently asked questions
Is the Word file editable?
Yes. The text is real text in a standard .docx file, not a picture, so you can edit, search and restyle it in Microsoft Word or any other app that opens .docx files.
Does it keep the formatting?
It keeps the structure: headings (as Word's Heading 1 and Heading 2), paragraphs, centered and right-aligned lines, indents and right-to-left text. It doesn't detect bold, italic, fonts, colors or tables, so the text uses Word's standard font.
Can I convert several images into one Word document?
Yes. Add as many images as you need; each becomes a page in the same file, in the order shown. You can reorder, turn or remove pages first. Every page of a multi-page TIFF is added.
Can I convert a scanned PDF to Word?
Not directly. Turn the PDF's pages into images with PDF to JPG, then add those images here.
Which languages can it read?
29, including English, Spanish, French, German, Arabic, Hindi, Urdu, Bengali, Chinese, Japanese and Korean. Choose the language in the image; you can add a second one for mixed text. Nothing is translated.
Does it work with handwriting?
Not reliably. The reading engine is made for printed text. Neat block capitals may work, but joined handwriting usually comes out wrong.
Why are some words highlighted?
With Highlight words to check turned on, words the engine wasn't sure about are marked in yellow, in the preview and in the Word file, so you know where to look first.
Are my images uploaded?
No. The text is read and the Word file is made in your browser. Only the reading engine and the language data are downloaded, the first time; your images and text never leave your device.
Can I use it on my phone?
Yes. Take a photo of a page or choose images from your gallery, then switch between the Image and Text tabs to check each page before you download.
How is this different from Image to Text?
Image to Text gives you the plain text of one image to copy or save as TXT. Image to Word makes a formatted Word document from one or many images, with headings and paragraphs.
Ready to turn your images into Word?
Add a photo, scan or screenshot above, check the preview and download your editable document.