🔎 Image to Text (OCR)
Extract text from images, screenshots and photos of documents — right in your browser, in 25+ languages. Your image is processed on your device and never uploaded.
Upload, take a photo, or paste an image
PNG, JPEG, WebP, BMP, GIF — up to 15MB. You can also drop a file here.
On-device OCR: recognition runs locally in your browser via WebAssembly. Your image is processed on your device and never uploaded; the OCR engine and language data are fetched once from tesseract's CDN and then cached. Works best on clear printed text — handwriting and cursive are not reliably recognized.
Extract text from images, screenshots and photos of documents using optical character recognition (OCR) that runs entirely in your browser, in 25+ languages. Upload a file, take a photo, or paste an image, let it recognize the text, then copy or download the result — your images and the extracted text never leave your device.
Frequently asked questions
How do I extract text from an image or photo?
Upload an image, take a photo with your camera, or paste from the clipboard, choose the language, and click to run OCR. The recognized text appears in a box you can copy or download. For photos of documents, turn on "Enhance for photos" to improve accuracy.
Can I take a picture of a document to extract the text?
Yes. On a phone, tap "Take Photo" to capture a document with your camera. Hold the camera steady, fill the frame with the text, and use good lighting. The "Enhance for photos" option applies grayscale and contrast, which helps a lot with camera photos.
Does it work on handwriting or old manuscripts?
It works well on clear printed text, including printed books and documents. Handwriting and cursive are not reliably recognized — the engine is trained on printed type, so a photo of handwriting will usually come out garbled.
Which languages are supported?
Over 25, including English, Spanish, French, German, Greek, Russian, Chinese, Japanese, Korean, Arabic, Hindi and more, plus bilingual combinations. Pick the language that matches your image before running OCR.
Is my image uploaded to a server?
No. The OCR engine runs locally in your browser using WebAssembly, so both the image and the recognized text stay on your device — safe for private documents. Only the engine and language files download once from a CDN.