Paul Lamere, who works for the Echo Nest and Spotify, has developed an algorithm that finds the drop in any song without even needing to look at the music.
He built it at MIDEM Hack Day - an event where coders from around the world come and build interesting music apps over the course of a weekend. "My hack is called The Drop Machine," Lamere wrote in a blog post. "It is a toy web app that plays nothing but the drops."
What's interesting is that the algorithm ignores the music entirely - instead, it looks at where people tend to scrub to in a song. "It turns out that the crowd knows exactly where the drop is," he wrote.
"When one person scrubs to a particular point in a song, we learn a tiny bit about how that person feels about that part of the song – perhaps they like it more than the part that they are skipping over."
I can feel it coming
In most tracks, therefore, it's just the most interesting bit that gets scrubbed to most often - the bit where the drums kick in in In The Air Tonight, for example. But when it comes to dance music - especially dubstep and brostep - there's a distinct peak about a minute in. "This is invariably 'the drop'," says Lamere.
Lamere's 'Drop Machine' searches through the most popular dance, dubstep and brostep tracks, therefore, and just pulls out the 10 seconds of each track centred around the drop. "The result is non-stop drop," he writes. Right now it only works internally at Spotify, but Lamere says that he's working on a public version. "Hopefully the world won't have to wait too long before you all can listen to the Drop Machine."
