Reviews for Screen OCR Translator
Screen OCR Translator by Alex
Review by Firefox user 20169910
Rated 5 out of 5
by Firefox user 20169910, 2 months agoThanks, I needed a free plugin that translates captions in images.
6 reviews
- Rated 5 out of 5by Firefox user 15479246, 8 days ago
- Rated 5 out of 5by lolcat, a month agoGreat extension, its the only one that works. If you could add an option to disable the overlay by default so I can only keep the translations on-hover, then it would be absolutely perfect.
Developer response
posted a month agoThanks! This option already exists. Go to Settings > Display > Initial overlay view and select "Original" from the dropdown. This will hide the translation overlay by default. - Rated 5 out of 5by Aphant, a month agoVery good. This add-on is great for reading manga and comics due to the overlaying of translated text. It really should be at the top of the results when searching for OCR add-ons. The only thing missing is being able to right click on the page or click on the toolbar icon and "translate all images on page".
Developer response
posted a month agoThanks for the kind words! I might add this feature in the future, thanks for the suggestion! - Rated 5 out of 5by Kamik, 2 months agoWorks great but could you add DeepL for translating? Also choosing different local OCR engines like Tesseract, Paddle OCR, Manga OCR, Baberu Ocr, Hayai OCR would be neat too
Developer response
posted 2 months agoThanks for the suggestions!
I'm considering adding DeepL support. The main issue is that their old API Free plan seems no longer available to new users, and the current one includes one million characters for free in total rather than per month, and after that you have to pay for a subscription.
Their ToS also explicitly prohibit the use of internal APIs, so I wouldn't want to add DeepL as a keyless integration. That means users would need to provide their own DeepL API key.
Would that still be useful to you?
As for alternative OCR engines, the extension bundles PaddleOCR because it provides a good balance of accuracy, language support, and can run directly in the browser. Tesseract can be faster, but it has been less accurate in my testing.
If there is a particular OCR engine or use case you're interested in, feel free to open GitHub issue.
Edit: DeepL support has been added in v0.19.0 - Rated 5 out of 5by Firefox user 19668205, 3 months ago