- Senator calls Facebook’s current election disinformation efforts ‘inadequate’ in letter 5 Years Ago
- The Phillie Phanatic mascot unveils a slimmer makeover 5 Years Ago
- YouTuber threatened with arrest after rapping about being a girl from Mecca 5 Years Ago
- Video shows flat-Earther ‘daredevil’ crashing to death after homemade rocket fails 5 Years Ago
- Cardi B defends Dwyane Wade’s daughter during Instagram Live 5 Years Ago
- YouTube briefly shuts down beloved ‘lofi hip hop radio’ channel, launching a new meme 5 Years Ago
- Neil deGrasse Tyson points out that Elsa from ‘Frozen’ has ‘horse-sized eyeballs’ Today 10:58 AM
- Republicans as Sanders rises: Watch out, we may vote for Trump Today 10:54 AM
- Amazon series ‘Hunters’ criticized by Auschwitz Memorial over fictionalized scene Today 10:45 AM
- Netflix’s ‘Gentefied’ is actually made for people of color Today 9:28 AM
- Drug dealer loses $60 million after misplacing his Bitcoin code Today 9:18 AM
- TSA bans employees from using TikTok Today 9:09 AM
- PewDiePie rips Jake Paul, calls his money-making venture ‘complete bullsh*t’ Today 9:05 AM
- Can a lawsuit in Illinois stop a dangerous new facial recognition app? Today 6:30 AM
- Fan uncovers ‘Westworld’ trailers hidden on fictional company’s website Sunday 8:18 PM
Hip-hop music’s highly dissected lyrics and deeply structured nature make it the perfect medium for computer-driven data research. Most rap songs contain three verses, all written in the same format: 16 lines of poetry, each a 4-beat measure in length.
Now, some researchers are parlaying their existing hordes of data about the metrics and patterns of rap into hot, original flows written by their computers.
You may remember Nordic fan extraordinaire Eric Malmi, the Ph.D. student from the University of Aalto in Finland who labored over countless hours of hip-hop music to create an algorithm that beat the genre’s most technical and accomplished rappers in a controversial ranking.
With the help of his colleagues, Malmi has taken the research behind that algorithm, which he named the Raplyzer, into a new realm. His new algorithm, dubbed DeepBeat, mines his database of rap music for similar lines across thousands of lyrics. It then composes a Sweet 16 of dope bars that rhyme, all sourced from thousands of different rap songs.
Malmi simply pulls one line from his 10,000-song, 100-artist-strong database, and before you can say “mayne, hold up,” a verse is generated, with an eye toward optimizing assonance rhymes. (These are the hard kind to make, because, beyond rhyming “party” with “Bacardi,” assonance poetry also rhymes as many parts of a given line as it can.)
“An 82 percent accuracy was achieved for separating the true next line from a randomly chosen line,” Malmi told Technology Review.
So is the resulting verse hot to death or straight trash? Judge for yourself.
For a chance at romance I would love to enhance /
But everything I love has turned to a tedious task/
One day we gonna / have to leave our love in the past/
I love my fans but no one ever puts a grasp/
I love you momma I love my momma – I love you momma /
And I would love to have a thing like you on my team you take care/
I love it when it’s sunny Sonny girl you could be my Cher/
I’m in a love affair I can’t share it ain’t fair Haha I’m just playin’ ladies / you know I love you.
I know my love is true and I know you love me too/
Girl I’m down for whatever cause my love is true/
This one goes to my man old dirty one love we be swigging brew/
My brother I love you / Be encouraged man And just know/
When you done let me know cause my love make you be like WHOA/
If I can’t do it for the love then do it I won’t/
All I know is I love you too much to walk away though.
Overall, it’s an incoherent narrative that fails at the art of storytellin’. Nonetheless, it’s a fun way see how Ghostface lines sync up with Eminem lyrics. And they are dense in structure, as Malmi pointed out: “DeepBeat outperforms the top human rappers by 21% in terms of length and frequency of the rhymes in the produced lyrics.”
Of course, that’s as far as the hype cycle for MC MacBook can go—until someone finds a way for it to perform.
Ramon Ramirez is the news director, and formerly the Dot's entertainment editor and evening editor. His work has appeared in the Washington Post, Grantland, Washington City Paper, Austin American-Statesman, and Austin Monitor.