This is the draft dataset that has been written by volunteers Nora Ash (Email: @gmail.com) for reason in here mind.
I am digitizing this data to make it accessible online.
The used model here is the TrOCR by MS researchers.
You can infer a notebook here:
