Skip to content
Devtoolry
Sign InJoin for Free

Image to Text (OCR)

Extract text from images, screenshots and scanned PDFs with free OCR in your browser. English, Hindi and 20 more languages; nothing is uploaded.

Runs in your browser · Nothing uploaded · No sign-up

Something not working, or missing a feature? Send feedback

About the Image to Text (OCR)

Image to Text reads the text in photos, screenshots and scanned PDFs with Tesseract OCR running in your browser, so you can copy it, edit it or save it as a .txt file. It reads English, Hindi and 20 more languages, including Bengali, Marathi, Tamil, Telugu, Arabic, Chinese and Japanese, and can read a second language and English in one pass. Scanned PDFs are read page by page. Nothing is uploaded.

Images are processed in your browser with its built-in codecs, so your photos and screenshots are never uploaded.

How to use the Image to Text (OCR)

  1. Choose the language of the text. For Hindi and other languages, keep Also read English on if the document mixes in English words or numbers.
  2. Drop images or a scanned PDF onto the box, choose files, or paste a screenshot with Ctrl+V anywhere on the page.
  3. Wait while each file is read; the first time, the OCR engine and the language data download once and are cached.
  4. Pick Lines as in the image or Join into paragraphs, fix anything that was misread, then copy the text or download it as a .txt file.

Common mistakes

  • Choosing the wrong language. Tesseract only recognises the scripts of the languages you choose. A Hindi page read as English comes out as nonsense; change the language and click Read again.
  • Trusting numbers without checking. OCR can confuse 0 and O, 1, l and I, or 5 and S. Check account numbers, amounts and dates against the original before you use them.
  • Photographing at an angle. Text that is skewed, curved (a book’s spine) or in perspective is read much worse. Hold the camera straight above the page, or crop and straighten the photo first.

Works well with the Image to Text (OCR)

Frequently asked questions

Is this OCR free, and is there a limit?

It is free with no sign-up and no daily limit, because the OCR runs on your own device. Scanned PDFs are read up to 50 pages at a time; split longer files and read them in parts.

Are my images uploaded?

No. Images and PDFs are read in your browser. Only the OCR engine (from this site) and the language data (from jsDelivr, once) are downloaded, which matters for ID cards, bank statements and contracts.

Can it read Hindi and other Indian languages?

Yes: Hindi, Bengali, Marathi, Tamil, Telugu, Gujarati, Kannada, Malayalam, Punjabi (Gurmukhi) and Urdu. Choose the language, and keep Also read English on for documents that mix in English words, names or numbers.

Can it read handwriting?

Only neat, printed-style handwriting, and not reliably. Tesseract is trained on printed text; cursive notes and forms filled in by hand usually come out with many errors.

How do I get the best results?

Use a sharp, well-lit, straight photo with the text filling most of the frame; crop away other things first. Screenshots work very well. Small images are enlarged before reading, but a blurry photo stays blurry.

Can I extract text from a scanned PDF?

Yes. Drop the PDF and each page is rendered at 300 dpi and read in turn, with page markers in the text. If the PDF already has selectable text, copying it from a PDF reader is faster and exact.

How do I keep the columns of a table?

Turn on Keep spacing (tables, code). Runs of spaces are kept, so columns stay roughly aligned in a monospace font. For anything more complex, check the numbers against the original.

Related tools