Spotify claims to develop an algorithm to sort songs by emotion, rather than genre: The human brain is wired to make connections with innumerable things that act as cues to trigger specific thoughts or emotions when we come across them. A specific aroma will take us back to our grandmother’s kitchen, and the sound of waves bring up memories of summers spent by the beach. Music is not only not an exception — we arguably form the most robust mental and emotional connections with musical cues. So for technology to tap into that to improve our musical (and hence, emotional) experience would be rather groundbreaking — and that’s exactly what the people at Spotify claim to have done, David Sumpter writes for The Economist’s sister magazine, 1843. They’re developing an algorithm that can apparently differentiate between a sad song and a happy one, thereby sorting mns of different songs into emotional categories to give users a better, more personalized listening experience. “The algorithm combines musical properties like volume, tempo and — most importantly — energy with “emotional valence,” a measure of how happy or sad a song makes you feel. Generally speaking, high valence sounds make people feel positive emotions, while low valence sounds are associated with negative emotions.” The development process relied on human feedback on their emotional reactions to songs that were either high or low valence sounds, which were then fed into a computer to allow it to learn the basis for analyzing the songs.

Tags: