Scanned Image to Text
OCR for scans and document photos — recognize text locally with WebAssembly. Clear, high-contrast scans work best.
Drop a file here
Best with clear scans — OCR runs in this tab. The Tesseract language pack is fetched from a public CDN.
How to use this tool
Add a scan
Photo or flatbed scan as JPG/PNG.
Run OCR
Text is recognized on-device — privacy-first.
Use the text
Copy into docs or download .txt.
From paper scan to TXT
Flatbed scans and phone photos of pages become searchable text without a cloud OCR vendor.
Scan quality tips
Aim for straight pages, even lighting, and 200–300 DPI equivalents. Blurry or skewed scans reduce accuracy.
Architecture
Image → browser → Tesseract WASM worker → recognized text → copy/download. No backend store.
Frequently Asked Questions
More on quality loss: conversion quality guides · compression guides
Is my scan uploaded?
No. OCR runs locally with Tesseract.js.
What about multi-page PDFs?
PDF OCR is planned later. Export pages to images, then OCR each page here.
Which language?
English is enabled now.
Does compressing or converting images reduce quality?
Lossy compression can. Use higher quality settings when fidelity matters, and keep an original master. Lossless workflows preserve pixels.
Can you recover quality after compressing?
Not from the compressed file alone. Keep the original upload/export and recompress from that master if you need a better result.
What is generational loss?
Quality damage from repeatedly re-encoding lossy images. Avoid save→compress→save loops; work from a master.
Ready to convert more formats?
Browse converters or jump back to the tool above.