PDF contains all text in images not able process for training and chat with PDF not working because of it.

Add OCR processing to recognise text within images.