Is my image or PDF uploaded anywhere?

No — recognition runs entirely on-device via a self-hosted OCR engine. The file, and the text extracted from it, never leave your browser.

Does it work offline?

Yes, once the OCR engine and your chosen language have loaded once, recognition works fully offline.

Can it turn a scan into a searchable PDF?

Yes — the Searchable PDF export embeds the recognized text as an invisible, selectable layer over the original scanned image.

Image to Text

Extract text from images and scans — offline.

Runs in your browser

Languages

Grayscale Auto threshold Upscale small images Invert

Confidence

Drop images or a PDF, or click to browse

PDF pages are rendered automatically (up to 20)

OCR runs entirely on your device via a self-hosted Tesseract engine — files are never uploaded. Approximate file sizes: 6.0 MB on first use per language set, cached after.

Compress PDF

Hash

About

Image to Text runs a self-hosted Tesseract OCR engine entirely in your browser to pull text out of photos, screenshots, and scanned PDF pages — the file itself never leaves your device, only the extracted words appear on screen. Preprocessing (grayscale, automatic black-and-white threshold, upscaling small images) runs before recognition to improve accuracy on low-quality scans, and low-confidence words are underlined so you know what to double-check. Six languages are supported, and a scanned PDF can be turned into a searchable PDF with the original text embedded, selectable and copyable.

Frequently asked

Is my image or PDF uploaded anywhere?: No — recognition runs entirely on-device via a self-hosted OCR engine. The file, and the text extracted from it, never leave your browser.
Does it work offline?: Yes, once the OCR engine and your chosen language have loaded once, recognition works fully offline.
Can it turn a scan into a searchable PDF?: Yes — the Searchable PDF export embeds the recognized text as an invisible, selectable layer over the original scanned image.