Local OCR – Screenshot & Image to Text

Turn text in screenshots, images, and multi-page PDFs into editable text without sending your content to a server. Local OCR runs its recognition engine, PDF renderer, and language models entirely inside Chrome. You can select part of the visible page, recognize a specifically right-clicked image, paste an image, or add images and PDFs from your device. Multiple files and PDF pages can be processed together. Built-in image adjustments can improve difficult scans. When the source has a clear layout, the extension can also create a structured result that preserves document organization. Recognized text remains editable before it is copied or downloaded. Twelve bundled language models are available, together with an automatic language option. Optional searchable history is stored only on the device and is disabled by default. The extension requires no account and contains no analytics, advertising, or tracking. It does not request persistent access to every website. Area capture cannot run on protected Chrome pages such as the Chrome Web Store and browser settings. Printed text generally produces better results than handwriting or highly stylized text.