How to Extract Text from an Image in Hindi and English (Free OCR)

Updated 2026-10-26 · 3 min read

Typing out text from a photo of a notice, a screenshot or a scanned page takes time and causes mistakes. OCR (optical character recognition) reads the text for you so you can copy and edit it.

The Image to Text tool reads English, Hindi and several other Indian languages. The text is read in your browser and your image is never uploaded.

Extract text from your image — Image to Text (OCR)

Step by step

  1. Add the image Drop an image into the box, click to choose it, or paste a screenshot with Ctrl+V. JPG, PNG and other common image formats work.
  2. Pick the language In 'Language in the image', choose English, Hindi or 'English + Hindi' for mixed text. Bengali, Tamil, Telugu, Marathi, Gujarati, Kannada, Malayalam, Punjabi and Urdu are also available.
  3. Extract the text Press 'Extract text'. The first time you use a language, its data is downloaded (a few MB) and then kept by your browser for next time.
  4. Check, copy or download The text appears in an editable box. Fix any wrong characters, then copy it or press 'Download .txt'.
  5. Many images? Use bulk Switch to 'Many files (bulk)', pick the language, add the images and press 'Extract text from all'. You can download each .txt file or all the text as one .txt.

Tips for better accuracy

  • Use a sharp, well-lit photo taken straight from above, not at an angle.
  • Crop the image to just the text before extracting.
  • Pick the right language. Hindi text read with English selected will come out as nonsense.
  • Use 'English + Hindi' only when the image really has both. A single language is usually more accurate.
  • Printed text works much better than handwriting.

Common uses

  • Copying text from a notice, circular or newspaper cutting.
  • Getting text from a screenshot of a chat or web page.
  • Reading text from a scanned PDF: first save the pages as JPG with the PDF to JPG tool, then extract the text here.
  • Turning a printed question paper or old typed letter into text you can edit and reprint.

Common Hindi OCR mistakes to check

Devanagari has matras, half letters and joined letters, so OCR makes a few typical mistakes. Read the result once and look especially for these:

  • Matras (like ि, ी, ु, ू) missing or swapped, especially in small or blurry text.
  • Half letters and joined letters (like क्ष, त्र, ज्ञ) read as separate letters.
  • The top line (shirorekha) cut off when the photo is cropped too tightly. Leave a little space around the text.
  • Numbers in Devanagari (१, २, ३) and English (1, 2, 3) mixed up. Check dates and amounts carefully.

Where to paste the text

  • Word or Google Docs, to edit and format it.
  • A WhatsApp message or email, to share it quickly.
  • Google Translate or another translator, if you need it in another language.

Questions

Can it read Hindi handwriting?

It is made for printed text. Very neat handwriting may partly work, but expect many errors.

Why are some words wrong?

OCR guesses each character. Blur, shadows, small fonts and the wrong language cause mistakes. Always read through the result.

Is my image uploaded?

No. Text recognition runs in your browser. Only the language data is downloaded the first time; your image stays on your device.

Can I convert a scanned PDF to text?

Yes, in two steps: save the PDF pages as images with PDF to JPG, then extract the text from them here using bulk mode.

Does it work offline?

After a language's data has been downloaded once, it is cached by your browser. You need an internet connection the first time.

Related guides