Table 5.

Medians of all audio features for the 6 clusters of the Sleep dataset.

		N	Danceability	Energy	Loudness (db)	Speechiness	Acousticness	Instrumentalness	Liveness	Valence	Tempo (bpm)
Cluster 1	Speechy Tracks	8275	0.68 (0.21)	0.45 (0.31)	− 10.01 (5.97)	0.33 (0.14)	0.50 (0.59)	< 0.01 (0.71)	0.12 (0.11)	0.46 (0.32)	102.0 (62.64)
Cluster 2	Radio Tracks	31,068	0.62 (0.20)	0.60 (0.22)	− 7.09 (3.62)	0.04 (0.04)	0.16 (0.28)	< 0.01 (0.08)	0.12 (0.08)	0.41 (0.32)	119.9 (46.1)
Cluster 3	Acoustic Radio Tracks	30,793	0.50 (0.20)	0.29 (0.19)	− 11.05 (5.5)	0.04 (0.01)	0.82 (0.23)	< 0.01 (0.007)	0.11 (0.04)	0.24 (0.2)	111.8 (46.4)
Cluster 4	Ambient Tracks	117,240	0.20 (0.15)	0.04 (0.10)	− 27.95 (10.4)	0.04 (0.01)	0.96 (0.10)	0.92 (0.09)	0.11 (0.03)	0.05 (0.08)	83.2 (50.4)
Cluster 5	Instrumental Tracks	32,736	0.66 (0.16)	0.18 (0.19)	− 17.44 (8.13)	0.05 (0.03)	0.89 (0.24)	0.89 (0.11)	0.11 (0.02)	0.40 (0.32)	92.5 (40.0)
Cluster 6	Live Tracks	5783	0.30 (0.21)	0.29 (0.38)	− 18.52 (15.4)	0.05 (0.03)	0.83 (0.47)	0.86 (0.88)	0.70 (0.26)	0.15 (0.26)	95.1 (43.0)