A few everyday examples of NLP include question-and-answering systems (such as Siri), machine translation (such as Google Translate), and chatbots and virtual assistants that are powered by NLP. You should now have a firmer grasp of some of the core features of natural language processing, including what is easy and what is difficult for machines, and how NLP is used in industry.
NLP has always been a favorite topic of mine !!
I will digress to circa 92 , where i wrote my 1st NLP/ domain specific interface. NB: (link is to a Facebook post). I was in the early research phase of subject/domain and predicate ecosystems. It was not a pure art then and its still not a pure art now. Today's dilemma still resides with subject vs predicate knowledge. - In short a lot has changed to driving a true NLP paradigm , but its all about how we derive at the outputs.
Having said that , I would still contend that today's NLP has to have "vernacular" speech. e.g French in Quebec is totally different to how its spoken in France, Likewise Russian in Ukraine is differently translated to Russian in Siberia. So the root question in NLP would be should this source sentence be classified as A (Quebecois) or B ( basque francophone) and should it be translated in "vernacular" ; X ( Quebec) or Y (basque french) .
How does the NLP engine determine "vernacular" speech and how to translate to another "vernacular" speech ??
Lets raise the bar.. Slang in one language into a another language ..