It may have been a few years, but no one has forgotten how awesome Dance Dance Revolution is. The innovative game allowed you to dance to some of your favorite songs, and led to some pretty epic Youtube videos.
And now robots have taken it over. For real.
StepMania, an open sourced version of the game, allows fans to make and share their own step charts for DDR. Of course, choreographing a new song can be intense and tedious. That’s why a group of researchers at the University of California, San Diego are working to create a way to speed up the process using computers and robots.
The team is building a neural network that automatically makes DDR-style step charts from raw audio recordings. Essentially, a neural network is a program that can learn specific tasks. Sort of like a computer brain. And now their computer brain is basically better and faster at creating DDR routines that humans.
Sigh. Robots win again.
The project started when researcher and DDR player Chris Donahue was trying to figure out a better way to extract data about music from recordings. This would lead to better routines for songs, created faster. As Donahue told The Outline:
Here’s a comparison between human choreography and their program, dubbed ‘Dance Dance Convolution’:
Another advantage to using this program is the ability to create more than one routine for a song. So you won’t get bored with the same choreography for your favorite song. And it helps with increasing or decreasing the difficulty of the movements.
Want to try it for yourself? They’ve made a demo of the tool available on their website. It looks like it’s time for us to break out that mat again!