Tokenizing tweets as ngrams in R

Question

I am trying to tokenize a collection of tweets as bigrams. I am using tidytext::unnest_tokens() function in R to do this. The function takes either token