Image to Text (OCR)
Turn screenshots, photos, and scans into editable text in your browser. Pick 26 languages or auto-detect the script, with dark-mode invert and contrast.
Still loading. If the tool doesn’t appear, reload this page.
About this tool
Image to Text (OCR)
Extract editable text from images, screenshots, and scanned documents - OCR in your browser, with 26 languages plus script auto-detect.
Image to Text (OCR) turns screenshots, photos of printed pages, receipts, and slides into clean, editable text directly inside your browser. Powered by WebAssembly optical character recognition, it reads your image in the browser without uploading it (the Tesseract engine and the language data download from public CDNs the first time), with optional contrast enhancement, dark-mode inversion, and interactive line-level bounding boxes. It downloads the Tesseract OCR engine and language data from cdn.jsdelivr.net on first use; your images stay in your browser.
How to use Image to Text (OCR)
- 1
Drop an image, choose a file (PNG, JPG, WebP, BMP, GIF), paste from clipboard with Ctrl+V, or click 'Try Sample' to test immediately.
- 2
Pick your document language and optional enhancements like High Contrast or Invert Dark Mode for dark screenshots.
- 3
Watch the progress bar while OCR extracts the text with confidence scores and interactive bounding boxes, then copy or download as TXT, Markdown, or JSON.
Why use Image to Text (OCR)
Private client-side OCR
Your images are read in your browser with WebAssembly and are not uploaded. The OCR engine and language data download from public CDNs on first use.
Smart image preprocessing
Toggle High Contrast binarization, Invert Dark Mode for dark-theme screenshots, or 90-degree rotation before scanning.
Interactive text boxes & cleanup
Inspect detected lines directly over the image, click any box to copy that line, or clean extra spaces and line breaks in one click.
Common use cases
- Copy code snippets, error messages, or UI labels from screenshots without retyping a single character.
- Digitize printed receipts, invoices, book pages, or handwritten-style flyers into searchable text.
- Extract Hindi, English, or multilingual text from social media graphics, slides, or PDFs.
Frequently asked questions
Are my images uploaded to a server?
No. Recognition runs in your browser with Tesseract (WebAssembly) and your image is not uploaded. On first use your browser downloads the Tesseract engine and the data for the language you pick from cdn.jsdelivr.net, then caches them.
How can I get better results on dark-mode screenshots?
Turn on 'Invert Dark Mode' in the toolbar. OCR engines are trained primarily on dark text over a light background, so inverting dark screenshots often improves the result.
Which languages are supported?
The language menu lists 26 languages, English + Hindi and English + Arabic pairs, and Auto-detect script. The languages include English, Hindi, Marathi, Bengali, Tamil, Arabic, Chinese (Simplified and Traditional), Japanese, Korean, Spanish, French, German, Thai, and Vietnamese. Each language downloads its own data on first use.
What does auto-detect do?
It looks at the image to find the script (Devanagari, Arabic, Han, Hangul, Cyrillic, Latin and so on) and then reads it with the matching language. It needs a few lines of text; if no script is found, or the text is in the Latin script, the image is read as English. Pick Spanish, French, German or another language yourself for accents.
Can I paste a screenshot directly?
Yes! Press Ctrl+V (or Cmd+V on Mac) anywhere on the tool page, or click the 'Paste' button to load an image straight from your clipboard.