In the scope of the Globalise project we've worked on automatic language identification of letters that were shipped back home by the VOC (East India Company) from 1610 to 1796, as preserved by the National Archive. They've been digitised, and automatically transcribed. While the vast majority is in Dutch (unsurprisingly), there are some notable and beautiful exceptions. My colleague has just published a blog post on it: https://globalise.huygens.knaw.nl/the-languages-of-globalise/

#globalise #humanities #history #voc #nlproc