‘Neural Karaoke’ Joins AI Robo-Rock Race

November 30th, 2016: As seen on Archive (PDF)

neuralkaraoke

Photo: Vimeo


Another week, another artificial intelligence lyric machine nipping at we the creative humans of the world.

Though this one’s not nearly as cool as Giorgio Cam by the team over at Google Creative Lab — mainly because there is no Giorgio Morodor hat tip — the ‘Neural Karaoke’ program by the University of Toronto is still at least just as much entertaining in the lyric realm.

Swapping a male voice in the ‘Giorgio’ set-up for a female one, ‘Neural Karaoke’, like ‘Giorgio Cam’ works off of image-recognition algorithms to churn out automated jams, associated said images with a bank of words programmed into its robot head to virtually freestyle.

The outcome is this Christmas monstrosity: [LISTEN]

Lots to decorate the room/The Christmas Tree is filled with flowers/I swear it’s Christmas Eve/I hope that is what you say/The best Christmas present in the world is a blessing/I’ve always been there for the rest of our lives

The rabbit hole goes deep on this project, with tales of 100 hours of online music and 50 hours of song lyrics making up the program’s 3390-word vocabulary, which then get strung at a rate of one word per beat.

Raquel Urtasun, an associate professor in machine learning and computer vision at Toronto’s computer science lab, explains via The Guardian, “We are used to thinking about AI for robotics and things like that. The question now is what can AI do for us? You can imagine having an AI channel on Pandora or Spotify that generates music, or takes people’s pictures and sings about them.”

We’ll spare the robot rule jokes for the fact that there are no damn flowers on that tree, ‘Neural Karaoke’ machine, hit the vocab books and maybe we’ll see ya at the jukebox hero contest some day.