Learn how to copy text from any image, screenshot, or scanned document with in-browser OCR, without uploading your files to any server.
Quick summary: OCR (optical character recognition) is the technology that turns text inside an image into text you can select, copy, and edit. You no longer need to retype what appears in a screenshot or a scanned document by hand. And the best part: you can do it directly in your browser, without installing anything and without your images ever leaving your device.
OCR stands for optical character recognition. It is a technology that analyzes an image, identifies the shapes of letters and numbers, and converts them into digital text you can select, copy, and paste. Put simply: you give it a photo containing text and it gives you that text back as if you had typed it yourself. It sounds like something only a large corporation would need to digitize archives, but in practice you run into everyday situations where you need it constantly:
Most online OCR tools work by uploading your image to a server for processing. That seems harmless, but think about the kind of images you usually need to run through OCR: screenshots of private conversations, photos of contracts, scanned medical documents, invoices with your tax details, bank statements. All of that ends up on a third-party server you have no control over. You do not know how long they store it, who has access, or what they do with that information.
Today it is possible to run OCR entirely in your browser thanks to Tesseract.js, a WebAssembly version of the most widely used OCR engine in the world. WebAssembly lets your browser execute high-performance code as if it were an installed application, but without installing anything. When you drag an image into the tool, it loads into your browser's memory, Tesseract.js analyzes it locally, identifies blocks of text, recognizes the characters, and returns the extracted text. Everything happens on your device. No data is sent to any server.
If the image has a lot of content surrounding the text you care about, cropping it first with the crop image tool will improve the result. And if you need to extract text from a scanned PDF, convert the pages to images first with the PDF to JPG tool.
It is worth being honest about the limitations. Modern OCR is excellent with standard printed text, but it struggles with handwriting (unless the handwriting is very clear and regular), highly decorative or stylized fonts, text on complex multi-colored backgrounds, and curved text like that on a can or a t-shirt. For these cases, OCR can give you a starting point that you then correct by hand, but do not expect a perfect result on the first try.
Yes, as long as the photo has sufficient resolution and the text is legible. Modern phone cameras are more than capable. What matters is that the photo is in focus and the text is not too small. If you can read the text when you zoom into the photo on your screen, the OCR engine will probably be able to read it too.
OCR works with images, not PDFs. But you can convert the pages of your PDF to images using the PDF to JPG tool and then run each image through OCR. It is an extra step, but it works perfectly for scanned PDFs, which are really just images wrapped in a PDF container.
No. All OCR processing runs in your browser via Tesseract.js (WebAssembly). Your image loads into your device's local memory, is analyzed right there, and the text is extracted without any data leaving your computer or phone. When you close the tab, no trace remains.
The engine supports over a hundred languages, including English, Spanish, French, German, Portuguese, and Italian. It also recognizes non-Latin alphabets such as Cyrillic, Greek, and Arabic. Selecting the language manually improves accuracy, although automatic detection works well for text in common languages.