Tiny OCR Autor: bipin

Tiny OCR lets users capture text directly from the visible part of the current Firefox tab.

How it works:

Click the add-on icon and start selection.
Draw a rectangular region on the page.
OCR runs locally on-device (offline).
Extracted text is shown in a result panel and can be copied to clipboard.
Key points:

Supports multiple languages/scripts (including Latin, Devanagari, Chinese, Japanese, Arabic, Cyrillic).
Designed for privacy: no account, no sign-in, no external OCR API calls.
Does not store personal records, browsing history, screenshots, or extracted text on remote servers.
Processing is done inside the extension runtime using bundled OCR models.
Limitations:

Works on the visible viewport of the current tab.
Accuracy depends on image quality, text size, and contrast.