Tool
Image to text
Extract printed English text, then edit, copy or download it.
Your image stays in your browser.No account, upload, watermark or analytics.
PrivacyEnable JavaScript to use this tool. Loading the local workspace when JavaScript is enabled. The answers below remain available without it.
FAQ
Questions about this tool
Are my images uploaded?
No. Images are processed in this browser. Using a free tool does not create a cloud capture. Text recognition downloads its software and English model when needed. Your image and recognized text stay in this browser.
Which image files can I use?
PNG, JPEG and WebP, up to 20 MiB per image, 40 million pixels and 16,384 pixels on either side. A low-memory device may need a smaller image. Recognition supports up to 16 million pixels. Select a smaller text area when the whole image is larger.
Why is processing taking longer?
Processing speed depends on the image and your device. Keep the tab open, or cancel and try a smaller image. The first recognition run also downloads the English model, which can take longer on a slow connection.
Which text can I extract?
The tool supports printed English. Small, stylized, handwritten or compressed text may be misread. Proofread names, numbers and identifiers.
Do I need to select an area?
No. Whole image reads the entire image by default. Choose Select area when you only need a passage or column, or when the image exceeds the recognition pixel limit.
What can I download?
Edit the recognized text, copy it or download a UTF-8 TXT file. The original image is preserved.
Why does the first extraction need a connection?
It downloads the recognition software and English model. That download is separate from your image, which stays on your device.