Quantifying rappers’ rhyming skills

“Women lie, men lie, numbers don’t” – Jay-Z

Last Monday night, I posted to my personal blog a small study where I analyse the lyrics of Finnish hip-hop artists. By Tuesday night, the post had attracted over 32,000 page views, several online newspapers had reported about it, and I had given an interview to a national radio station. In this post, I will give a brief summary of what all the fuss was about.

Multi-syllable rhymes

In rap lyrics, assonance, where words don’t have the same ending, but they share a vowel sound, is the most typical form of rhyming nowadays [1]. In multi-syllable rhymes (multis), it is not only the last syllable but multiple syllables that share a vowel sound. For example:

“This is a job – I get paid to sling some raps,
What you made last year was less than my income tax [2]

As one author puts it: “Multis are hallmarks of all the dopest flows, and all the best rappers use them” [2].


In an attempt to compare the rhyming skills of Finnish rappers, I wrote a program called Raplyzer (Raplysaattori in Finnish). Raplyzer automatically detects rhymes from lyrics which is relatively straightforward in the Finnish language since it is pronounced as it is written. This allows us to detect rhymes by discarding all consonants and finding matching vowel sequences. Skipping some details, we can simply

  • Find all matching vowel sequences
  • Compute the sum of their lengths, and
  • Normalize by the number of words in the lyrics.

This gives us an average rhyme length (I call it rhyme factor) for each song. Averaging over all the songs of an artist, we obtain a rhyme factor for each artist, which I then used to rank the most popular Finnish rappers.

Raplyzer also measures the size of each artist’s vocabulary by counting the number of unique words among the latest 9,000 words in the lyrics of the artist. A similar analysis of rappers’ vocabularies has been done previously for English speaking rappers [3].

Why it went viral?

Finally, I list a couple of things I believe contributed to the popularity of the post.

  • I intentionally published the analysis right after Cheek (one of my study subjects and perhaps the most popular music artist in Finland nowadays) had played at the Helsinki Olympic Stadium in two concerts which were both sold-out within about an hour. There were also lots of other news articles around him at the time.
  • Rappers have a tendency to boast about their lyrical skills so I think people want to know whether there is any scientific evidence to support the claims of various rappers.
  • Social media. When I posted the link to my blog post on Monday night in Facebook, it was shared 13 times. By Tuesday noon a relatively popular Finnish rapper had somehow discovered the story (it would be cool if I could somehow see how the information had propagated and reached him in Fb!) and posted it on his own Facebook wall. This new post was then shared 85 times giving the story a much bigger momentum.

[1] http://genius.com/posts/24-Rap-genius-university-rhyme-types
[2] http://www.flocabulary.com/multies/
[3] http://rappers.mdaniels.com.s3-website-us-east-1.amazonaws.com/

Posted by Eric

This entry was posted in data science and tagged , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *