In 2016 the most interesting and prolific lyricist in the genre of American roots music is a twitter bot named horse_bluegrass.
Programmer Jared Wenerd fed the lyrics of 1,796 bluegrass songs into a text prediction algorithm. The algorithm creates sentences with a certain degree of randomness, but using predictions of words likely to follow the preceeding word, based on the input of the original songs fed into the program. The end result are lyrics that are at times nonsensical, but at other times quite poignant and profound.
The code takes text, parses it into individual words, to create a model where the algorithm knows the likeliness that one word will follow another or end a phrase. For instance starting with the word “in” it knows that a likely word to follow will be “the”, “a”, or 43 other different words. The algorithm decides to go with “the” due to the statistical likeliness and randomness. It then continues and chooses the next word after “the” using the same process… and so on until the algorithm decides the phrase should end. Once it has a complete phrase, it publishes the text to Twitter
.
The twitter account updates every couple of hours.
Here’s some of horse_bluegrass’ fine work. Certainly as good as, if not better than, anything coming out of Nashville in 2016. Check it out, no songs about pick-up trucks or beer:
Via: Medium