Variance in Features: Original Songs vs. Remixes
The art of remixing music has been around for as long recorded music itself. Experimental remixing even led to the creation and establishment of hip-hop as a genre. Today, remixes often frequent Billboard charts, especially with the explosive popularity of EDM (electronic dance music) within the last decade. As music sharing platforms, such as SoundCloud, and audio software become more and more accessible, I want to take a look at how remixes typically differ from their originals by analyzing their features.
Dataset
I created two playlists: p1, where I added 201 original songs, and p2, where I added a corresponding remix for each song in p1.
The original songs span across a range of genres including pop, hip hop, R&B, EDM, rap, rock, alt-rock, soul, and country.
Exploratory Data Analysis (EDA)
Even before delving into further understand differences, it is clear from the playlists lengths that the remixes are, on average, longer. The p1 playlist is 12 hours and 5 minutes long, whereas the p2 playlist is 12 hours and 44 minutes long.
I used Python to obtain 2 csv files for each of the playlists.
The features I considered are as follows:-
- Beats Per Minute (BPM): The tempo of the song.
- Energy: The energy of a song — the higher the value, the more energetic the song.
- Danceability: The higher the value, the easier it is to dance to this song.
- Loudness: The higher the value, the louder the song.
- Valence: The higher the value, the more positive mood for the song.
- Acoustic: The higher the value the more acoustic the song is.
- Popularity: The higher the value the more popular the song is.
Using R, I created various plots to visualize the differences and similarities for each feature.
Further, I created a divergent bars to illustrate the differences more clearly to determine which of them were significant.
It is thus clear that the remixes of songs are:-
- More Energetic
- Have More Beats per Minute
- Less Acoustic
and happen to be
4. Less Popular
