How to extract text from an image
- Drop the image on the page, click to choose it, or copy an image or screenshot and press
Ctrl+V. You can add several images or a multi-page PDF at once. - Pick the language of the text. For Indian bills and documents, “English + Hindi” (or another regional language) reads both scripts.
- Wait a few seconds while the text is read. The first time, the recognition engine and language data (a few MB) are downloaded; after that they come from your browser's cache.
- Check and correct the text in the box if needed, then copy it or download it as a PDF or TXT file.
Compare the text of two images
Comparing two pictures by eye is slow and easy to get wrong, especially with long bills, invoices, price lists, contracts or forms. In Compare two images mode, drop one image on each side. The text of both is extracted and compared line by line:
- Changed lines are shown side by side, with the exact words that differ highlighted.
- Lines that appear in only one image are marked in red or green.
- Numbers and amounts that changed, such as quantities, prices, taxes and totals, get their own table, so a different total on two bills stands out immediately.
- Download the comparison as a PDF report to share, including the full text of both images.
OCR is never perfect, so a misread character can show up as a difference. You can correct the extracted text in either box and the comparison updates. To compare text you already have, use the text diff tool.
Tips for better results
- Sharp, straight and well lit: hold the phone parallel to the page, avoid shadows and glare, and fill the frame with the document.
- Keep Enhance photo on for phone photos; it turns the image into high-contrast grey. For clean screenshots it makes little difference.
- For bills and receipts, set Layout to “Bill / receipt” so lines are read top to bottom as one column.
- Text PDFs (exported from software, not scanned) are read directly with no OCR at all, so the text is exact.
- Handwriting is not supported well; printed text works best.
Private by design
The recognition engine (Tesseract) runs inside your browser, so your images and the extracted text are never uploaded to a server. That makes it safe for bills, bank statements, ID cards and other personal documents. Only the engine and language files are downloaded from a public CDN the first time.
Frequently asked questions
How do I convert an image to text?
Drop the image on this page, or paste it with Ctrl+V. The text is recognised in your browser in a few seconds, and you can copy it or download it as a PDF or TXT file.
Can I compare two bills or invoices?
Yes. Choose “Compare two images”, drop one bill on each side, and you get the lines that differ side by side, plus a table of the numbers and amounts that changed, such as quantities, rates and totals. Download it as a PDF report.
Can it read text from a PDF?
Yes. Text PDFs are read directly and exactly. Scanned PDFs are rendered and run through OCR, up to 15 pages at a time.
Does it support Hindi and other Indian languages?
Yes. Choose English + Hindi, Marathi, Bengali, Tamil, Telugu, Gujarati, Kannada, Punjabi or Urdu. Many other languages are available too, including Spanish, French, German, Arabic, Chinese and Japanese.
How do I save the extracted text as a PDF?
Click “Download PDF”. A print window opens; choose “Save as PDF” as the destination. The PDF contains the text in a clean, searchable layout.
Are my images uploaded?
No. Text recognition runs entirely in your browser. Your images and their text never leave your device.