The future we saw in sci-fi cinema has already entered reality with artificial intelligence. From Alan Turing training machines to decode the Enigma during World War 2 to AI producing music in virtual realities, artificial intelligence has come a long way. Machine learning has given computers the ability to learn without being explicitly programmed. It seems quite magical to the common eye, but really it’s just mathematics.
Galileo Galilei said “Mathematics is the language with which God has written the universe.” Music is more than just a collection of notes which create harmony. It is about rhythm and melody, and the changing of notes in relation to time. Interestingly, arithmetic and geometric patterns can be found in music and its compositions if examined closely. Mathematics can be seen in music through Fibonacci Music.
Mathematics and Music: Fibonacci Music
The Fibonacci Sequence is a mathematical theory which constructs an infinite series of integers. It begins with 0, followed by 1 and each successive term is constructed by adding the two previous terms. The calculation will be made as 0+1=1, 1+1=2, 2+1=3, 3+2=5, and so on. Thus, the sequence would look like 0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377, 610, 987, 1597, 2584 and so on.
The figure to the left implies that dividing a line into segments according to the Fibonacci ratio, implies that the ratio of the length of a to a+b is the same as the ratio of the length of b to a. These ratios converge to a constant limit which is called the golden ratio. The golden ratio is an irrational number which is defined as ψ = 1.61803398…
The golden ratio is meant to make objects aesthetically pleasing. It is found in nature abundantly, from sunflower seeds to snail shells. In music, the golden section is often used to generate rhythmic change or to develop a melody line, and is found in the musical timing of compositions. The climax of a song, for example, is often found at the point of the golden ratio (approximately 61.8% of the way through a composition). Deliberate application of the golden ratio can be seen in the musical genius Mozart’s work too. There's a note of mathematical equations in the margins of some of his compositions like Fantasia and Fugue. Debussy, Stockhausen, Bartók, Stravinsky, Manzoni, Ligeti, are also some of the best known names in the music industry, who have voluntarily chosen the use of these magical proportions.
Figure 2: The Fibonacci Spiral
Can artificial intelligence use the application of mathematics to produce music?
Historically, yes. The first computer generated music was played in 1951 on a gigantic contraption built by the British computer scientist Alan Turing. However, the audio produced was distorted and the frequency dispersion was inaccurate. But after tweaking the it’s speed, compensating for a “wobble” in the recording and filtering out extraneous noise, the beautiful sound produced by the computer opened many gates of potential AI music.
Taryn Southern is one of the first modern day pioneers producing music through artificial intelligence. She produced ‘I AM AI’, a 2018 album featuring eight tracks produced entirely by AI. Digital avatar Miquela Sousa, better known as Lil Miquela, is a computer generated teenager Instagram model and influencer. The co-founder of Brud Technologies: Trevor McFedries, turned Miquela into a virtual popstar, and released her first AI generated single ‘Not Mine’ in 2017. Since then she’s released more than 10 non-album singles over the course of four years.
Music produced by artificial intelligence is gaining fast paced recognition and competition. In July 2021, a world-wide AI Song Contest featuring 38 independant music compositions portrayed the collaborative effort of musicians and AI researchers. All these tracks can be heard here: https://www.aisongcontest.com/participants-shuffle
On 6th April, 2021, Artificial Intelligence released a “new” Nirvana song, 27 years after the death of the lead singer Kurt Cobian. It was created by feeding MIDI (Musical Instrument Digital Interface) files into Google's AI program Magenta, which learns how to compose in the style of given artists by analysing their past work. Vocals for the song are provided by Atlanta musician Eric Hogan, frontman of Nirvana tribute act Nevermind: The Ultimate Tribute to Nirvana. You can listen to this song here: https://youtu.be/qf6eOSJgN0Y
This AI released music created a huge whim amongst Nirvana fans, some arguing that it’s ‘insulting to the artist’ and the ‘song lacks soul.’ But who decides if it is so? AI produced music has as much soul into it as any other song. Each track is the result of AI programs analysing up to 30 songs by each artist and granularly studying the tracks’ vocal melodies, chord changes, guitar riffs and solos, drum patterns, and lyrics to analyse what their “new” compositions would sound like. Researchers believe that listening to music triggers opioids—the body’s natural pain relievers. Relaxing music can alleviate stress by lowering cortisol levels, which is the hormone released in response to stress. If musical composition, regardless of it being produced by a human or a machine, has the capability to make us feel, it is music. Besides, if the title said that this was an unreleased Nirvana song, one would have probably believed it.
It is natural to be concerned about artificial intelligence taking over the music industry. However, as quoted by Eric Hogan, “There’s an inordinate amount of human hands at the beginning, middle and end to create something like this… 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 just to get to a point where a song is listenable is actually quite significant.” Websites like LyricJam are considered morally dubious because they are a system that generates lyrics just by hearing the instrumental composition of the artist. We fail to recognize that these systems were introduced for the purpose of inspiring the artist's own creativity by suggesting fresh new ideas and expressions that can spark an idea, not do the work for them. Technological advancement is the inevitable future. Many ambitious AI projects like N-Synth, Douglas Eck’s Magenta and AIVA are the future of music. But legally, who is the real artist behind the machine composed art?
As an artistic work, music may be protected by copyright and related or neighbouring rights. While copyright protects original works, i.e., the musical composition or the lyrics, in the case of music, neighbouring rights relate to the performance or interpretation, or to the fixation made for the sound recording. Countries like UK, South Africa, Hong Kong, India, Ireland, and New Zealand have entitled ownership machine produced music to either to the programmer or to the user of the producing computer program.
The creation of music can still be heavily influenced by humans, even if done with the assistance of an AI system. A product can be considered original and copy-rightable when a significant amount of creative human effort has been applied. Nonetheless, it cannot be disregarded that in the future, AI may be able to compose songs autonomously with minimal or insignificant human involvement.