Image & PDF OCR (Text Recognition)
Extract text from photos, scans and PDFs — 11 languages, copy or download as TXT
Server-side FFmpeg handles any format — including HEVC/H.265 files your browser can't play. Files are deleted from the server within 30 minutes.
⚠️ Please don't close or refresh this page until it's done, or you may lose the result.
💡 Large files travel at your connection's pace: the upload continues on its own and resumes safely, so a slow link only affects time, never your credits.
How the recognition pipeline works
Both modes share a server-side OCR engine, pre-tuned differently:
- Photo mode — deskews tilted pages and normalizes exposure before recognition; best for documents, signs and handwriting attempts.
- Screenshot mode — preserves character spacing for interface text: chat bubbles, menus, terminal output and pixel fonts survive where general OCR smears them.
- Languages — 100+ via Tesseract with automatic script detection; mixed Chinese-English captures handled per line.
- Output — plain text preserving reading order, plus confidence markers on low-certainty words you should proofread.
Two OCR engines, one workspace
Photo mode reads camera captures with perspective correction; screenshot mode is tuned for chat UI, code and game text — pick per job without leaving the page.
Two recognition modes
General image OCR and screenshot-focused OCR in one tool.
Multi-language support
The engine recognises text in all eleven site languages.
Reads 11 languages
Uploaded images are removed from the server after processing.
Confidence flags
Low-certainty words are marked so you know exactly which spots to proofread instead of trusting blind output.
How it works
- 1
Choose Image or Screenshot mode.
- 2
Upload the picture containing the text.
- 3
Copy or download the recognised text.
- 4
Copy the recognized text inline or download it as UTF-8 TXT.
Frequently asked questions
Can I OCR a PDF?⌄
Yes, both modes accept PDF pages as input.
Does it support handwriting?⌄
It works best on printed text. Handwriting recognition is not guaranteed.
Is my image kept?⌄
No. It is deleted from the server after the text is extracted.
Why does handwriting often fail?⌄
The engine targets printed text. Neat block letters work sometimes; cursive handwriting needs a dedicated handwriting model that this tool intentionally does not pretend to have.
Related tools
What is Image OCR?
Image OCR turns pictures of text into editable, selectable text. It works on scanned documents, photos of pages, and screenshots of apps or chat conversations.
What you can do
- Extract text from photos and scans
- Recognise text in screenshots and chat windows
- Process multiple languages with the same engine
When to use it
- Digitising printed notes or receipts
- Copying text from images you cannot select
- Turning screenshots into editable content
Images are uploaded for server-side recognition and then deleted.