Closed fznx922 closed 1 year ago
Hello. and thanks for the kind words.
It sounds as if the language detection fails.
Can you tell me your source and target language? I would guess Source is "Auto" in your case because otherwise it should not try to load the language detection model.
Not sure though what that has to do with char encoding, 🤔
Also does it work if you set the source language to the actual language of the scanned text?
My guess is that it might fail because of a character encoding of the scanned text. So maybe it would also be helpful if you could provide me with an example text it fails with so i can reproduce it?
thank you in advance
thank you so much for the quick reply! so currently i have the source lang as Japanese, and source of translation is Japanese, target English
when i configure and press process window with ocr. i get the prompt saying its loading, then when i go to the log window it shows me the same 2% progress and the error, attached a screenshot as i couldn't see an openable log file?
thank you so much for your time!
and for additional information the translation function and also the speech to text and caption is all working well, absolutely amazing ! just the ocr component im having difficulty with :(
little bit of an update, i was able to bypass the error by manually downloading craft_mlt_25k.pth and japanese_g2.pth so now it gets past the error, but now im recieving an error on the log saying "charmap: codec cant encode characters in position 26-30: character maps to
hopefully this helps to find a solution, not sure if its just unique to what im doing, but all im trying to currently do is run a bit of a dirty translation on a switch game.
Thank you. i was able to reproduce it. It happened when the window title contained unicode characters. 🙈 Will be fixed in the next version.
Should be solved in the new update i just released.
I close this issue. If you still have any issue or a question, feel free to open a new one.
hey Sharrnah
Love the application, works amazing in real time, on the OCR side of things im trying to get it to work but come across the issues when i try to start it
it says its dowloading detection model, then freezes on 2% and is saying cant encode character u2588 in pos. but then it gets cut off and cant read the rest
are you able to help or know of a solution?
thanks so much!