Where is the Drama takes any song input recognized by Spotify and analyses it to find the 30 seconds or so of highest drama, defined as the portion of the song with the largest increase in volume.
Loudness in audio is a somewhat tricky concept that’s correlated with average level, but doesn’t map directly. I’m assuming he’s using the same definition of loudness as this.
The app grabs the detailed audio analysis for the song from The Echo Nest. This includes a detailed loudness map of the song. This is the data I use to find the drama. To do so, I look for the part of the song with the largest rise in volume over the course of a 30 second window (longer songs can have a bit of a longer dramatic window). I give extra weight to crescendos that culminate in louder peaks (so if there are two crescendos that are 20dB in range but one ends at 5dB louder, it will win). Once I identify the most dynamic part of a song, I pad it a bit (so we get to hear a bit of the drop after the build).
There’s more details here and here, but the proof as they say is in the pudding. As one of the commenters says, it correctly identifies the drum fill from ‘In the Air Tonight’ and the beginning of the guitar solo in ‘Stairway to Heaven’ so I’m going to go ahead and call it a success.
(Also in the hackiest of ingenious hacks it uses a play 4’33” command as a stop button)