Can you see music?
This question sparked the research behind my master's thesis, carried out together with a friend with a background in music, to understand how a song is actually built (structure, rhythm, form) before even trying to translate it into image.
Songviz is the tool that came out of this research: it lets you listen to music while watching it at the same time. The system runs on two independent variables: shape indicates which instrument is playing, and colour indicates which chord, a circle for percussion and keyboards, lines for strings, typography for voice; each chord, in turn, is translated into a colour calculated mathematically from the relationship between sound frequency and the wavelength of visible light, not chosen by feel.
Every graphic element is anchored to a precise sonic event in the score: if the kick drum hits on beats one and three of every four, a circle appears on screen on beats one and three, in the colour of the current chord. Rhythm, structure and form of the song become a system of visual signs that moves in real time with the listening.
The starting point was the work of Luigi Veronesi, who in 1972 was the first to propose a sound-to-colour transposition on a scientific basis rather than through subjective association, applying to the visible light spectrum the same mathematical ratios that link the notes of a musical scale. I built on this principle, correcting a few calculation errors present in the original.
To build the colour scale, I took a reference octave (523–1046 Hz) and applied its mathematical ratio to the visible light spectrum (380–760 nm, ENEA data), obtaining a precise colour for each note, then extended it to the semitones, for a total of twelve colours instead of the seven of the diatonic scale.
The colour of the song's key thus becomes the main colour of the composition. For the other chords I applied the rules of colour theory (complementary and analogous), assigning each degree of the scale (dominant, subdominant, leading tone...) a colour in precise relation to that of the tonic, instead of choosing them by feel.
The system was applied to two songs at opposite extremes, to show it holds up across opposite genres: Kiss by Prince, funk, in a red/orange palette, and Henna by Lucio Dalla, a ballad, in a blue palette. Same visual grammar, completely different outcomes.
Vertical 9:16 format, designed to be watched on Spotify, Apple Music and social media while listening, with a hypothesized extension to live concerts too, from synchronized LED walls to the audience's light-up wristbands.
The project closed out my master's degree, completing the research into the relationship between the visual and the sonic, with a positive response at the thesis defense.