How to specify langauges for each pdf seperately in new version?

[aumungray]

Hello,

Thanks for the great library.

In last version we had parameter called metadata-files which took a json file as an input that had mapping of language(s) for each pdf file. In the new version this parameter is not there and the config takes the language in comma separated form. What I have observed is the OCR sucks for few pdf where the right language is not specified.
Any solution to specify language for each pdf in bulk in new version.?
Also in surya is there an inbuilt mechanism which detects the language of the page?

Thanks