Convert an image to text free with OCR that runs in your browser. Copy text from JPG, PNG, or WebP photos and screenshots. Nothing is uploaded.
Processing stays in your browser.
Add a JPG, PNG, or WebP image, or paste a screenshot. English text is read on your device and never uploaded.
Reading the whole image.
Drag over the text you need to read only that part. With a keyboard, focus the image, then use the arrow keys to move the area and Shift with the arrow keys to resize it.
Practical guide
This image to text converter reads printed English text from a photo, screenshot, or scanned page and gives you plain text you can copy or save as a .txt file. Recognition runs inside your browser, so letters, IDs, and receipts never leave your device.
Many people look for image to text offline because the photo is private. Here the image and the text never leave your browser. The OCR engine and English text data, up to about 7 MB, download from this site the first time you extract text, and your browser can reuse them later. No account and no page limit.
Shoot the page flat, in even light, and fill the frame with the text. Drag a box over the paragraph you need so headers, stamps, and backgrounds are skipped. Small screenshots are enlarged automatically, and the confidence note tells you how much proofreading to expect.
The output is plain text only. Columns, tables, fonts, and page layout are not kept, so this is not a JPG to Word layout converter. Handwriting is not reliable. For a scanned PDF, convert its pages with PDF to JPG first, then read each page image here.
Practical answers about privacy, formats, and output quality.
No. The image is decoded, cropped, and read by an OCR engine running inside your browser. The OCR engine and English text data are downloaded from RealScaner itself, and neither the image nor the extracted text is sent anywhere.
Your image is always read on your device, but the page needs a connection the first time to download up to about 7 MB of OCR files from this site. Later runs can reuse the copy in your browser cache, which is faster. A full offline mode is not promised.
English only for now. Text in other languages or scripts will come out wrong or empty. More languages are planned and will be added as separate downloads so English users do not pay for them.
No. The result is plain text in reading order. Columns, tables, fonts, and images are not kept, and handwriting is usually not recognized reliably. Proofread names, numbers, and amounts before you use the text.
Convert the scanned pages with PDF to JPG first, then open each page image here. For best results use 200 DPI or higher and crop to the part of the page you need.
Accuracy depends on the photo. Blur, shadows, curved pages, tiny text, busy backgrounds, and tilted shots all lower accuracy. Retake the photo flat and well lit, or drag a box around just the text you need, then extract again.