Can I turn a JPG into text?

Yes. Drop the JPG in above and click Extract text — you get the words back in an editable box, ready to copy or save as a .txt file. PNG, WebP, BMP, GIF and TIFF work the same way.

Does it work on a photo of a printed page?

Yes, and that is one of the things it is best at. Photograph the page as straight-on and as evenly lit as you can — the flatter and sharper the text, the cleaner the result.

Does it read handwriting?

Not this tool — it is trained on printed and typed characters, and on real handwriting it returns mostly noise. There is a separate Handwriting OCR tool for that, linked below: it runs a model built for handwriting, cursive included, and gets roughly 60–70% of the words right, which is a draft you correct rather than a finished document. It costs a one-time ~250 MB download and about a minute a page — which is exactly why it is a separate tool and not the default here.

What kind of text works best?

Clear, reasonably straight printed or typed text — documents, screenshots, signs, receipts. Very low-resolution, skewed or heavily patterned backgrounds are harder. For handwriting, use the Handwriting OCR tool linked below instead.

Is my image uploaded anywhere?

No. The text recognition runs entirely in your browser on your own device. Your image is never uploaded, and nothing is kept once you close the tab.

Is it really free? Are there limits?

Yes — completely free, no account, and no limits. Because everything runs on your device, there is nothing for us to meter.

Which engine does it use?

Tesseract (Apache-2.0), compiled to WebAssembly and served from our own origin. It is excellent on printed and typed text. For handwriting, use the handwriting OCR tool instead — it runs a much larger transformer model.

Image to text

Pull editable text out of a picture, photo, screenshot or scan, then copy it or download it as a .txt file. It all runs on your device.

Your image
Drop an image here — or click to choose
JPGPNGWEBPTIFF
Your files are never uploaded — everything runs on your device.
Text

How to turn a picture into text

Add your image

Drag and drop a photo, screenshot or scan (JPG, PNG, WebP), or click to choose one. It loads straight into your browser — nothing is uploaded.
Extract the text

Click Extract text. The characters are recognized on your device — the first run downloads a small English model, then it is cached.
Edit, copy or download

Fix any stray characters in the editable box, then copy the text or download it as a .txt file.

What you get

Editable result

The recognized text lands in an editable box with a confidence score, so you can tidy it up before you copy.

Photos, screenshots, scans

Works on printed and typed text in photos, screenshots, documents and receipts. JPG, PNG, WebP, BMP, GIF and TIFF are all supported.

Private and free

No accounts, no page limits, no watermarks. Your image is never uploaded — the whole thing runs in your browser.

The engine behind it

Powered by Tesseract, the long-established open-source OCR engine, compiled to WebAssembly. About 5 MB and roughly a second a page — but it is trained on printed characters, so it cannot read handwriting.

Frequently asked questions

Can I turn a JPG into text?: Yes. Drop the JPG in above and click Extract text — you get the words back in an editable box, ready to copy or save as a .txt file. PNG, WebP, BMP, GIF and TIFF work the same way.
Does it work on a photo of a printed page?: Yes, and that is one of the things it is best at. Photograph the page as straight-on and as evenly lit as you can — the flatter and sharper the text, the cleaner the result.
Does it read handwriting?: Not this tool — it is trained on printed and typed characters, and on real handwriting it returns mostly noise. There is a separate Handwriting OCR tool for that, linked below: it runs a model built for handwriting, cursive included, and gets roughly 60–70% of the words right, which is a draft you correct rather than a finished document. It costs a one-time ~250 MB download and about a minute a page — which is exactly why it is a separate tool and not the default here.
What kind of text works best?: Clear, reasonably straight printed or typed text — documents, screenshots, signs, receipts. Very low-resolution, skewed or heavily patterned backgrounds are harder. For handwriting, use the Handwriting OCR tool linked below instead.
Is my image uploaded anywhere?: No. The text recognition runs entirely in your browser on your own device. Your image is never uploaded, and nothing is kept once you close the tab.
Is it really free? Are there limits?: Yes — completely free, no account, and no limits. Because everything runs on your device, there is nothing for us to meter.
Which engine does it use?: Tesseract (Apache-2.0), compiled to WebAssembly and served from our own origin. It is excellent on printed and typed text. For handwriting, use the handwriting OCR tool instead — it runs a much larger transformer model.

Wondering why some images read perfectly and others come out as nonsense? What OCR is, and the four things that break it. Handwritten instead of printed? This tool cannot read it — try handwriting OCR, which uses a model built for handwriting and returns a draft you correct. Working from a screenshot? Try screenshot to text.

Your image

Text