Technology Review showcases DeepBeat, a computer algorithm that is able to outperform human rappers by 21% in "terms of length and frequency of the rhymes".
The highly structured nature of rap makes it particularly amenable to computer analysis. And that raises an interesting question: if computers can analyze rap lyrics, can they also generate them?
Today, we get an affirmative answer thanks to the work of Eric Malmi at the University of Aalto in Finland and few pals. These guys have trained a machine-learning algorithm to recognize the salient features of a few lines of rap and then choose another line that rhymes in the same way on the same topic. The result is an algorithm that produces rap lyrics that rival human-generated ones for their complexity of rhyme.
|Hipsters (Are Roaming the Campus)|
|The Daft Punk's Console|
|"Women are more attracted to a man if he's holding a guitar."|
|YEARS: a Record Player that Plays Slices of Wood|
|Time Travelling Radio|
|"The world has lost its last shot at staving off dangerous global warming."|
|How to Build a 4 Barrelled Rocket Launcher|
|"I made a self driving potato."|
|Schools Monitoring Students' Social Media Feeds|
|WannaCry is Childsplay Compared to This|
|"They don't want to believe that their own intelligence is deterministic."|
|“Drone carrying defibrillators could begin operating in Sweden.”|
|"Is meant to produce more energy than it consumes."|
|“Making paper out of stone.”|
|“It’s not something you would expect to see there and not something we’ve seen there before.”|
|A Simple Task for CRISPR|
|“We have to forget steel as a core employer.”|