Drop images here to extract their text
Drag & drop, , or paste a screenshot with Ctrl+V.
PNG · JPG · WEBP · BMP · GIF · PDF, as many as you like.
100% private: images are read in your browser and never uploaded.
No text yet. Add an image and press Recognize. The extracted text shows up here, ready to edit and copy.
OCR (optical character recognition) turns a picture of text into text you can select, edit and search. Drop in a photo of a page, a screenshot, a scanned letter or a PDF, press Recognize, and the words appear in an editable box next to the image, ready to copy into an e-mail, a document or a spreadsheet.
The recognition runs entirely in your browser with Tesseract, the open-source OCR engine. Your images and PDFs are never uploaded: the page only downloads the OCR engine and the language model for each language you pick, and your browser keeps the language model for next time. There is no account, no sign-up and no daily limit.
It suits anyone who needs text out of a picture: students copying a quote from a photographed book, office workers retyping a scanned invoice or contract, developers grabbing an error message from a screenshot, or anyone digitising printed notes in Czech, English and ten more languages.
Press Add images, drop files on the image area, or paste a screenshot with Ctrl+V (Cmd+V on a Mac). Each image or PDF page appears as a thumbnail in the queue.
Open Languages and tick every language the text is written in. English is selected by default. Pick only the languages that really appear in the text.
The first run downloads the language model and the status bar shows the progress. Several images are read one after another, and the button turns into Stop if you want to end the batch early.
Correct any mistakes in the text box, use the highlighter to find uncertain words, then press Copy or Download. On a phone, switch between the Image and Text tabs.
Yes. It is free to use, with no account, no sign-up and no daily limit.
No. Every image and PDF is read inside your browser. The only downloads are the OCR engine and the language model for each language you pick, both from a public CDN, and the language model is then stored in your browser. Your images and the recognized text are not sent anywhere.
English, Czech, German, Spanish, French, Italian, Dutch, Polish, Portuguese, Russian, Slovak and Ukrainian. You can select several at once, for example English and Czech for a bilingual document.
Yes. Add a PDF and each page becomes its own item in the queue, up to the first 20 pages of a file. The pages are rendered as images and read with OCR, so this also works for scanned PDFs that have no text layer.
PNG, JPG, WEBP, BMP and GIF images and PDF files. HEIC photos from an iPhone and TIFF scans are not accepted here: convert them to JPG or PNG first, for example with the TechTools File Converter.
Only partly. The engine is built mainly for printed text, so neat block capitals may work, but joined-up handwriting usually comes out with many mistakes. Check the highlighted words and correct them in the text box.
It is the engine's own estimate, from 0 to 100 %, of how sure it is about the text it read. Words below 70 % can be highlighted with the highlighter button above the text and outlined in red on the image with the boxes button, so you know where to look for mistakes.
No. Results live only in this browser tab and are gone when you reload or close it, so copy or download what you need. Only your settings (recognition languages, layout mode, B&W and interface language) are remembered in this browser.
Yes. Press Add images to choose a photo, then switch between the Image and Text tabs. The first recognition downloads the language model, so use Wi-Fi if your mobile data is limited.