X

News and Innovation from Oracle UK and Ireland

Blue Jeans and Bloody Tears

By John Abel

The Eurovision Song Contest on Saturday night promises a spectacle of flamboyant acts and decided unique music from across Europe, all coupled with the hilariously acerbic commentary from Graham Norton.  Everyone will be tuned into their television sets to see which performer will be crowned the winner for their undoubtedly catchy song, which we likely all enjoy as a secret, guilty pleasure.

With the excitement building, Nimrod Shapira approached us with the idea to make a Eurovision styled song – something that honoured the fun and merriment of contest, but also something that could, at least in theory win.

With the excitement building, Nimrod Shapira approached us with the idea to make a Eurovision styled song – something that honoured the fun and merriment of contest, but also something that could, at least in theory win.

To do this team team decided to test the limits AI in creative processes and have our AI Cloud in collaboration with other technology experts help write the song for us. Rather than just have the song written by AI we wanted to demonstrate the role humans bring in augmenting creative processes, rather than being replaced by AI.

We teamed up group of Israeli artists, including new media artist Eran Hadas, and musicians and who fed 200 previous Eurovision entries into a machine learning cloud system to create melodies. Eran combined his poetry and AI skills to write the words.  Then composer-songwriter Amir Sheinfeld selected the most suitable parts to create a song that celebrates Eurovision – its melodrama, kitsch and summer camp ambiance, its humour and its gimmicks.

The result? A duet on disillusioned love sung by Cohen but co-created by a robot, Blue Jeans and Bloody Tears

Be warned: it is a little catchy and you might end up humming it in the office.

 

 

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.