A song by the name “Blue Jeans and Bloody Tears” has been algorithmically generated by a deep learning cloud-based AI which was fed thousands of Eurovision songs and lyrics for reference.
The result, however, is quite fascinating as human and machines jam together to having a happy-miserable existence. It’s pretty surreal to think about the meaning of the song.
The song was compiled by a group of programmers, artists and musicians who wanted to challenge and test the limits of human creativity through the AI. In turn, they created the ultimate Eurovision anthem which they describe as “containing camp, kitsch, melodrama, gimmicks and humour of Eurovision songs.”
The Making Of The Song
The project was helmed by Israeli digital artist Nimrod Shapira who was a former creative director at interactive multimedia video creation company Eko. In an interview, he stated that he couldn’t pass on an opportunity where he can test the limits of human creativity by combining it with machine learning. The idea, however, came into being when Israel won the Eurovision last year and is getting ready to host the 2019 edition at Tel Aviv.
The song was revealed on Oracle Cloud Day conference in Airport city.
The song is described as “disillusioned in love” is a duet between an AI and Izhar Cohen who won Israel their first Eurovision in 1978.
The deep learning algorithm produced the music which was later assembled by Amir Scheinfield who was the produced of ‘Toy’ which one last years Eurovision winner.
The questions posed by the song?
The quality of the song, however, poses some serious questions about the versatility of deep learning AIs and the limits of human creativity itself. The song captures the actual formulaic nature of Eurovision and sounds pretty nice. Is human creativity not as absurd as we think it is? Is AI the future of creativity?
Music, being one of the most human expressions there is. Isn’t as human anymore?
Give the song a shot yourself and comment your thoughts below!