Eko & Vinda Folio — Spotifyr for Therapy
Musical analysis in R package ‘spotifyr’
Few days ago (as for 16 December 2019) a Georgian duo Eko and Vinda Folio ended their tour throughout France, bringing their debut album Therapy to European audiences. To add to the celebratory mood for their success I analysed their album with R spotifyr package that connects developer version of Spotify to R’s robust data analytic and visualisation tools.
Spotifyr gathers statistics of the tracks and gives you features of the tracks as their danceability, loudness, keys used in compositions, musical modality, tempo, valence, acoustic quality of the songs and many more. Descriptions of the algorithms are not always revealing about their underlying principles but they give you a general sense of the included elements.
The compositions are either written in major or minor musical modality, “the type of scale from which its melodic content is derived.” The keys tell you which key is mostly used in a composition with its modality. “He Was All of Them” is written in F sharp minor; “Emotional Captivity” in G major. The graph on the left shows all the modes and keys of each song in Therapy. If you know these chords on any instrument, you might get the feeling of these songs prior to listening to them, as keys are related to certain emotions. Therapy (C minor) has to be sadder than Ramble Around (G major).
Eko and Vinda Folio played their gigs in France along with their favourite band ‘Motorama’ whose sound is not that far from that of Eko and Vinda. Spotify can tell us which one was more likely to get the crowd bouncing with its danceability feature. I took their albums respectively (Eko and Vinda’s Therapy and Motorama’s Many Nights) and pitted them against Joy Division’s classic album Unknown Pleasures to give a bit wider perspective to the comparison.
Density plots show you how the data points are distributed — whether they are concentrated at some level or spread out rather evenly. For example, Unknown Pleasures has many tracks that are hard to dance on and others that are suitable for dancing. Most of the tracks in Therapy are high in danceability index while Many Nights slightly trails behind it with its double hunch distribution.
According to Spotify tracks are more energetic when they are fast, loud and noisy. Listening to Therapy makes you feel energetic with some help from morning coffee, but not so much as Black Sabbath would do, which would fill all the bars in the graph. Overall it aligns more with an active mood when outside in the streets, taking a stroll in a sunny weather.
The tracks also differ in terms of valence which Spotify explains as “a musical positiveness conveyed by a track.” This musical positiveness means that tracks that are high on this index are associated more with positive emotions as happiness and cheerfulness while on the other extreme songs are more depressing or angry in their sound. One might think that ‘sad’ tracks would have lower beats per minute, in other words, they would be slower, but these three compositions shown on the graph are quite fast, actually faster than some of the ‘happy’ songs on the upper right end. Subjective experience of listening to the album bears with these numbers as most of the tracks have a positive vibe.
Now give the algorithms in your brain a chance to validate its artificial colleague’s competence and listen to the album on Spotify or watch a video of the song ‘Therapy’.
