search

gali8 / Tesseract-OCR-iOS

Tesseract OCR iOS is a Framework for iOS7+, compiled also for armv7s and arm64.
http://www.nexor.it
MIT License
4.22k stars 949 forks source link

Can you use alternative languages from Tesseract? #193

Closed losh11 closed 9 years ago

losh11 commented 9 years ago

I had read here before that using custom training files for other languages fails. Not too sure.

kevincon commented 9 years ago

Although there have been several reports from users that using custom language files doesn't work, I think generally it's because either they trained the file with a mismatched version of Tesseract from this repo (this repo uses Tesseract 3.03-rc1) or they're trying to use a different engine like cube that is known not to work with user-created language files.

I have been successful in creating a custom language file and using it with this library, so let us know if you have an issue with your own custom language file and we'll help you out.

dawand commented 9 years ago

Hello Kevin,

Is the trained data files generated for 3.02 work for this version of Tesseract? I tried Arabic and Turkish files for 3.02 downloaded from Google code and they do not work! Any help would be appreciated

Dawan

kevincon commented 9 years ago

@dawanjiali Can you open a separate issue for your question and provide the specific error message you're getting? Also double-check that your trained data files work with Tesseract 3.03-rc1 on your desktop. If it doesn't work with that version on desktop, it won't work with this library either.