What to select
Select one clear JPEG, PNG or WebP image containing printed English text.
Image OCR
Extract English text from JPEG, PNG and WebP images privately in your browser.
We use reasonable safeguards to help protect your files and information while you use the service.
High-confidence browser OCR. The image is enhanced and checked across multiple recognition passes. Results are shown when reliability reaches 80%.
Supported formats
Accepts one JPEG, PNG or WebP image and returns editable plain text. Printed English text is supported in this first version.
Limits and safety
Images are limited to 10 MB, 10,000 pixels on either side and 25 megapixels. OCR uses your device, so large images can take longer on older hardware.
Frequently asked questions
Is my image uploaded?
No. Recognition runs in a temporary browser worker. The OCR software and English model are served as cacheable website assets; your image and extracted text stay on your device.
How can I improve accuracy?
Use a sharp, upright image with large printed text, strong contrast and little background clutter. The 80% reliability gate compares multiple OCR passes, but it is a quality signal rather than a guarantee that every character is correct.