Image to text converter (OCR)
Turn a photo, screenshot or scanned PDF into text you can edit and copy. Reads English, Hindi, or both together. The text recognition runs inside your browser, so your files are never uploaded.
Everything runs on your device. Nothing you add or type is sent to any server.
How to use it
- 01
Drop JPG, PNG or WebP images, or a scanned PDF, into the box above. You can add several files; they are read in order.
- 02
Choose the language of the text: English, Hindi, or English + Hindi for documents that mix both.
- 03
Press Extract text and wait while each page is read. Progress is shown page by page.
- 04
Check the result, correct any mistakes in the box, then copy it or download it as a .txt file.
What it works well on
Optical character recognition (OCR) works best on clear, printed text: typed letters, book pages, notices, receipts, forms, screenshots and documents scanned with a phone scanning app. A sharp, well-lit photo taken straight on gives far better results than a blurry or angled one.
Accuracy depends on the scan. Small or faint print, shadows, folds, decorative fonts, tables and pages with several columns can produce mistakes or text in the wrong order. Handwriting is unreliable and often comes out wrong. Always read the result before you use it, especially names, numbers and amounts.
Hindi and mixed documents
Choose Hindi for text written in Devanagari. Many Indian documents mix Hindi and English, such as forms, bills and government notices; for those, choose English + Hindi so both scripts are read. Expect a few errors with matras and joined letters, especially in small print.
Scanned PDFs and privacy
For a PDF, each page is turned into an image at 300 DPI and then read, so it also works as a free OCR PDF tool, up to 50 pages at a time. If a PDF already contains selectable text, copying it from a PDF reader or using PDF to Word is more exact. The OCR engine (Tesseract) and its language data are downloaded from this site the first time you extract text, then run on your device. Nothing is sent to a server.
Frequently asked questions
Is it really free and private?
Yes. There is no sign-up and no limit on how often you use it. Images and PDFs are processed in your browser and never uploaded.
Why is the first run slower?
The first time, your browser downloads the OCR engine and the language data, up to about 8.5 MB depending on the language. After that they come from the browser's cache and start much faster.
Can it read handwriting?
Not reliably. It is made for printed text. Neat block capitals sometimes work, but joined-up handwriting usually comes out wrong.
Will the result keep the layout of my page?
No. You get plain text with line breaks. Tables, columns, bold text and images are not kept, so tidy the result in the box before copying.