LISTEN: AI Composes Songs Based On Music From Nirvana, Amy Winehouse & Hendrix & It Sounds Creepily Similar
The Lost Tapes of the 27 Club project was launched recently. This project uses AI software to create songs in tribute to the style of musicians who died at the age of 27, popularly known as the 27 Club.
The project features songs composed by AI that sounds creepily cool and similar to the original style of Nirvana, Amy Winehouse, Jimi Hendrix and The Doors.
As explained in Rolling Stone‘s feature, Google AI’s program Magenta was used to analyse the musician’s music style to create the instrumental tracks. An artificial neural network was then used to generate the lyrics while the vocals were still sung by real people.
Here’s a track featured on the mixtape project titled ‘Drowned in the Sun’, which sounds like a really cool Nirvana track with vocals by Eric Hogan, frontman of Atlanta Nirvana tribute band.
And if you’re wondering how the other tracks based on the music of The Doors, Amy Winehouse and Jimi Hendrix sound like, have a listen below:
The Doors – ‘The Roads Are Alive’
Amy Winehouse – ‘Man, I Know’
Jimi Hendrix – ‘You’re Gonna Kill Me’
Magenta was given 30 of each artists’ songs as MIDI files and the program goes through a complex process of analysing and identifying to create new compositions that make the tracks listenable and even good.
Lost Tapes of the 27 Club is a project from Over The Bridge, a non-profit organisation based in Toronto, Canada that supports members in the music industry who struggle with mental illness.
Besides highlighting the importance of mental health issues, Over The Bridge hopes the project showcases how much work actually goes into creating AI music and that it will never replace the real thing.
“There’s an inordinate amount of human hands at the beginning, middle and end to create something like this,” explained Michale Scriven, a representative from Lemmon Entertainment, whose CEO is on Over The Bridge’s board of directors.
“A lot of people may think [AI] is going to replace musicians at some point, but at this point, the number of humans that are required to get to a point where a song is listenable is actually quite significant,” he added.