Image to Text (OCR)

Runs in your browserNo sign-up requiredNo watermark

Drop photos, screenshots or scanned PDFs, or click to choose

You can also paste a screenshot with Ctrl+V.

Language in the image (up to 3)

Clean up photo first (enlarge and boost contrast)

The text will appear here, ready to edit, copy or save.

What Is OCR?

Optical Character Recognition (OCR) reads the text visible in an image and converts it into actual, editable, copyable text — useful for pulling a quote out of a screenshot, digitising a printed page or scanned PDF, or grabbing text from a photo of a notice, letter or form without retyping it. This tool runs the open-source Tesseract engine in your browser, supports Hindi and 11 other Indian languages, and cleans up and straightens each page before reading it.

How to Use This Tool

1Add photos, screenshots or a scanned PDFSeveral at once is fine; every PDF page is read (up to 30). You can also paste a screenshot with Ctrl+V.
2Pick the languageChoose up to three, for example English + Hindi for a bilingual form. The reading data for a language downloads once and is then remembered.
3Extract the textEach page is cleaned up (enlarged, greyed and contrast-boosted), straightened automatically and read. A score shows how sure the reader is.
4Fix, copy or downloadEdit the text in the box, join broken lines into paragraphs, then copy it or save a .txt file.

Common Reasons People Reach for OCR

Pulling a quote or a paragraph out of a screenshot instead of retyping it by hand.
Digitizing a printed page, an old document, or handwritten-looking printed notes into searchable, editable text.
Extracting text from a photo of a whiteboard, sign, or receipt taken on the go.
Copying text out of an image where the original source file (a PDF, a design file) isn't available anymore.

Good to Know

A sharp, well-lit, straight-on photo gives the best results. Fill the frame with the text and avoid shadows.
For a document with two scripts, such as an Aadhaar card or a bilingual form, select both languages.
If the confidence score is below 60%, retake the photo closer and in better light, or turn off "Clean up photo" for very clean screenshots.
Always check names, numbers and dates before using the text: OCR can confuse similar-looking characters.
Your files never leave your device. Only the language reading data is downloaded, the first time you use each language.

Frequently Asked Questions

Why did the first extraction take longer than expected?

The OCR engine downloads a small trained-data file for your chosen language the first time you use it (a few megabytes) — after that, it's cached in your browser and subsequent extractions are much faster.

How accurate is the extracted text?

It depends heavily on image quality — clean, high-contrast, well-lit text recognizes very well, while handwriting, unusual fonts, or low-resolution photos will have more errors. Always review the result.

Can it read handwriting?

It is made for printed and typed text. Very neat handwriting in capital letters sometimes works, but joined-up handwriting usually doesn't.

What languages are supported?

English, Hindi, Gujarati, Marathi, Bengali, Tamil, Telugu, Kannada, Malayalam, Punjabi, Odia, Urdu and Sanskrit, plus Arabic, French, German, Spanish, Chinese and Japanese. Pick up to three when an image mixes languages, such as English and Hindi.

Can I edit the extracted text before downloading it?

Yes — the result appears in an editable text box, so you can fix any recognition errors directly before copying it or downloading it as a .txt file.

Can I convert a scanned PDF to text?

Yes. Add the PDF and each page (up to 30) is turned into an image and read, with the text of all pages in one box. For PDFs that already contain selectable text, copying it directly is faster and more accurate.

Is my image uploaded to a server?

No. Text recognition runs entirely in your browser, so your images and PDFs never leave your device. Only the language reading data is downloaded.