If anyone has worked rather extensively with Abbyy for OCR with romanized Sanskrit texts with diacritics and is happy to share their supplementary recognition file resulted from the training of Abbyy on such texts we would be most grateful. 

We are planning to OCR a number of texts and it would be great not to have to train Abbyy from 0 through I must say the latest versions are very good at recognizing diacritics if you simply type a sample of the character into the additional character set option in the software.

There is reference to this planning to be made available in this post in 2010 but thereafter I have not seen further in this regard.

FYI: "You can save the trained pattern via menu TOOLS > OPTIONS > tab READ > "User patterns and languages"."

Kindest Wishes,

Martin Gluckman