Music has evolved along with technological advancement, a true testament to new forms of adoption that human’s desire. In the beginning, there was analog music, played in acoustic instruments and recorded in vinyl. As the synthesizer replaced the piano, and guitarists preferred electric over acoustic ones – we entered the digital age of music. Next, the explosion of electronic music took over the world, as producers abandoned instruments to create sounds out of a computer. So, what does the next era of music entail?
It is indeed, music created by seeking the help of Artificial Intelligence (AI). Composer and producer Benoît Carré, along with musician and technology researcher François Pachet, have collaborated with Artificial Intelligence to form the group Skygge. Translating to “shadow” in Danish, Skygge was inspired by the Hans Christian Andersen story of the same name. The group has released their first album, “Hello World”, the first AI-human collaborated album in the world.
Although AI has been proven to exhibit racist and sexist biases, just like a majority of humans, it’s still unclear how AI perceives the progression of tunes that make noise sound music to human ears. A piece of pop music requires backstories and a creative process, something that AI algorithms lack. Nonetheless, their ability to generate unfamiliar and novel sounds make them a welcome addition to any musical group. AI finds inspiration in the social and musical experiences of thousands of lives and draws on that output to offer new melodies, instrumentations, and other musical elements, based on statistical probabilities in a dataset.
In 2017, Dadabots (led by producer Zack Zukowski and technologist CJ Carr) produced the first neural-network-created heavy metal album, Coditany of Timeness. Even as early as 1980’s, researcher and musician David Cope explored algorithmic composition with the creation of his Experiments in Musical Intelligence (EMI) program. In 1993, Cope released the Bach-inspired Bach by Design album using EMI. EMI’s music has mostly been appreciated on technical instead of artistic merit. At that point, it seemed A.I. music functioned best with predictable parameters, like the predominantly rules-based music format of Bach’s fugues.
While producing Hello World, each Skygge artist and producer interpreted the Hans Christian Andersen fable within a chosen genre and worked in conjunction with the A.I. technology. The European Research Council funded Skygge to explore AI in pop music production. Instead of using neural networks, as done in Google DeepMind’s Deep Dream Generator, the producers used Sony’s Flow-Machines tools to generate Markov Chains to create catchy tracks. Neural networks require a substantial amount of information to produce an outcome, while Markov chains have the advantage of being able to produce statistical models from much smaller databases. Based on information acquired from previously recorded music, Flow-Machines suggested melodies, accompaniments, and instrumentation that producers could accept, alter or reject – creating the Human-AI collaboration.
Seeking the assistance of AI in pop music can push the boundaries of familiarity into new territories. Pop-singer Kiesza, one of the contributors to Hello World, created the melody for her track “Hello Shadow” using Flow-Machines. Kiesza said: “This melody sounded different from anything that I’d actually ever heard, I loved it from the beginning. Even though it’s still really haunting, it’s still really catchy.”