CJK(Chinese-Japanese-Korean) OCR seems to be working from studio but not in du-app.
Issue Description
In this scenario, the Document Understanding process is run by using Invoices ML Extractor to extract data, and the UseServiceSideOCR option is set as true. The connection is done via Automation Cloud for AI Center.
In the ML invoice package, if UiPath Chinese-Japanese-Korean OCR is used with the below properties, a 500 error always occurs.
Error:
![]()
(OCR Engine settings in ML Package of Cloud)
Resolution
The UiPath Chinese, Japanese, Korean OCR will be deprecated starting in January 2025. We recommend using the UiPath Extended Languages OCR instead.
