Image to Text (OCR)

Extract text from images—fast, private, and 100% local to your browser.

Upload Image

Drop, select, or paste an image. Large, sharp images work best.

Preview Your image preview will appear here.

Hold Ctrl/⌘ to select multiple.
Try 6/4 for documents; 7/8 for snippets.

Binarise to boost contrast (0=dark, 255=light).
Preview updates live; OCR uses this processed version.
0%
—

Extracted Text

Tips for Better OCR

  • Use the Upscale slider (200–300%) and Threshold for crisp, high-contrast text.
  • Pick the closest language(s) to your document (e.g., eng+fra).
  • Try PSM 6 for documents or PSM 7 for a single line.
  • Enable Orientation detection or set the Rotate slider manually.

Frequently Asked Questions

No—OCR runs fully in your browser. Only the language data file is fetched from a CDN the first time, then cached by your browser.

Your browser downloads the WebAssembly core and language file once. After that, it’s much faster and works offline.

Yes—use Download TSV for structured results (words with confidence and positions).