Screenshot OCR
Screenshot OCR is useful for text in images, video subtitles, games, and apps that do not allow copying. It opens the OCR panel and keeps the recognized source text available for editing. If you only want the translation, use Screenshot Translation instead.
Start Screenshot OCR
Section titled “Start Screenshot OCR”By default, Screenshot OCR appears in the translation panel header. You can customize this area and add the button again if it was removed. See Customize Action Positions.
You can also choose OCR → Screenshot OCR from the tray menu without opening the translation panel first. If you use it often, assign a key combination to Screenshot OCR on the Shortcuts page.
Recognize text on the screen
Section titled “Recognize text on the screen”- Start Screenshot OCR from the translation panel, tray menu, or a shortcut.
- Drag to select a tight region around the text you need.
- Wait for the OCR panel to show the screenshot and recognized result.
- Review and edit the text, then copy, translate, or continue processing it.
Work with the result
Section titled “Work with the result”Edit or recognize again
Section titled “Edit or recognize again”You can edit the recognized text directly. Use Text tools to remove blank lines, merge lines, trim each line, or remove all spaces.
If the result is inaccurate, switch the OCR capability or recognition mode and choose Recognize again. Float Translate processes the same screenshot, so you do not need to select the region again.
Copy or translate
Section titled “Copy or translate”Choose Copy to copy the text after your edits. Choose Translate to send the current text to the translation panel.
To copy every completed recognition result automatically, open the More menu in the OCR panel and turn on Auto copy.
Inspect regions and QR codes
Section titled “Inspect regions and QR codes”OCR capabilities that provide source coordinates can show recognition regions. Turn on Show recognition regions from the More menu, then click a region to copy that line.
When the screenshot contains a QR code, the OCR panel shows its content separately. You can copy the content or open it in your browser when it is a link.
If recognition does not finish
Section titled “If recognition does not finish”| What you see | What to do next |
|---|---|
| No OCR capability is available | Add a service. If a service already exists, make sure its OCR capability is enabled and configured, then try again. |
| No text or QR code is recognized | Select a clearer, tighter region, or switch the OCR capability or recognition mode and recognize again. If you use Windows built-in OCR, also confirm that the Windows OCR language required by the source text is installed and selected. |
| Recognition fails or times out | Select a smaller area and retry. For online OCR, also check your network connection. |
| A system permission prompt appears | Follow the in-app instructions to grant access, then start this screenshot recognition again. |
| Protected content cannot be captured | Screenshot OCR cannot bypass content protection enforced by the target app or operating system. Use a copy or export method allowed by that app. |
For other issues that you cannot resolve here, see Troubleshooting.