[Question] Is there a way to change the OCR model?

[kouyakamada]

I would like to OCR a PDF document written in Japanese. The accuracy of the layout recognition is great, but the OCR is useless.
So I researched the OCR engine PaddleOCR and found that there is a model for Japanese(japan_PP-OCRv3_rec).
I would like to replace the OCR engine, but how do I set it up?

[myhloli]

We will support select ocr language at next version.

4b372f3