Image to text (OCR)
Free, private image to text OCR that runs in your browser. Extract text from a screenshot, scan, or photo, then copy it or download as .txt. No upload.
100% private — files never leave your deviceRuns OCR in your browser. The first run downloads the English OCR model (~10–15 MB) once, then it's cached.
Working…
How to use
- Add your image. Drag and drop or select a screenshot, scan, or photo in PNG, JPG, or WebP format.
- Run the recognition. Start the OCR and wait a few seconds while the text is detected on your device.
- Check the text. Review the extracted text and fix any characters that came out wrong, especially on low-contrast images.
- Copy or download. Copy the result to your clipboard or save it as a clean .txt file.
Features
- 100% in your browser. Powered by Tesseract.js, your image is read locally and never leaves your device.
- Selectable, editable text. Turn pixels into real text you can highlight, copy, edit, and reuse anywhere.
- Copy or download as .txt. Send the result straight to your clipboard or export it as a plain-text file.
- Faster after the first run. The English language model is cached on first use, so repeat scans skip the download.
- Free with no limits or sign-up. Scan as many images as you like with no account, watermark, or hidden paywall.
When to use it
- Copy text from a screenshot of a chat, error message, or social post you can't select
- Digitize a scanned document, receipt, or printed page into editable text
- Grab a quote or paragraph from a photo of a book, sign, or poster
- Pull text out of an infographic or slide image for notes or citations
- Extract a code, address, or reference number from a product photo
Is it private?
Your image is processed entirely in your browser with Tesseract.js and is never uploaded to any server.
Frequently asked questions
Is this image to text tool really free?
Yes. It's completely free with no account, no watermark, and no usage limits. You can scan as many images as you want.
Is my image uploaded anywhere?
No. All OCR runs locally in your browser with Tesseract.js. Your image and the extracted text are never sent to a server.
How does the text extraction work?
The tool runs the open-source Tesseract.js OCR engine on your device. It analyzes the shapes in your image and converts recognized characters into selectable text, all without an upload.
How accurate is it?
Accuracy depends on image clarity. Clear, high-contrast, straight printed text reads very well, while blurry photos, handwriting, unusual fonts, or skewed angles can produce mistakes, so it's worth a quick proofread.
Why is the first run slower?
The first time you use it, your browser downloads the English OCR language model once (roughly 10-15MB from a CDN). It's cached afterward, so later scans don't repeat that download.
What image formats are supported?
Standard image files like PNG, JPG, and WebP all work. Cropping the image to just the text area usually improves both speed and accuracy.
Can it read handwriting or other languages?
It's tuned for printed English text and handles that best. Handwriting is unreliable, and recognition of other languages or non-Latin scripts may be limited without the matching language data.
Last updated