When I try to transcribe the Devnagari script book in multilingual Wikisource ( https://wikisource.org/), the OCR is not working.
You can check the below link
https://wikisource.org/w/index.php?title=Page:एका_विचाराची_जिवीत-कथा.pdf/81&action=edit&redlink=1, and click the "Transcribe text"
Unfortunately, Multilingual Wikisource doesn't make it easy for the OCR script to send the right language to the OCR engine.
Leaving out the language (via the 'Advanced Options' form), gives better results, e.g. see here.
I think the fix here is to a) not send 'en' as the language for any pages on Multilingual Wikisource; and b) work on T279405, so there's a way to choose which language is used.
In the meantime, a workaround is to use the 'Advanced Options'.