How to read two language content in one image using OCR activities

Help Studio
, ,
1

I have a pdf with some images where it has English and Arabic characters in same page and need the content of image using OCR.
When I give language in OCR as English, Arabic characters are also converting into English , and getting some unwanted characters in that string variable. And same occurs in Arabic also.
How to get both language string from OCR.

Can anyone help me with the solution?

2

Hi there - I believe you need to use OCR twice for the same page where one model will extract English data and other one Arabic data as I don’t think one model can handle both. We are using this approach with Sanad.ai (Sanad AI OCR and DU - RPA Component | UiPath Marketplace | Overview) especially for Arabic data extraction