What should I do if I want to replace the OCR model

[Lobskodax]

Hi, @VikParuchuri . I found that the Surya model has poor recognition rate for Chinese characters and Chinese layout. If I want to replace it with another model, how should I do?

[VikParuchuri]

Can you send me an example pdf? You would have to modify the code to change the layout model. For the OCR, you can use tesseract (see the README).