←back to thread

Shared DNA in Music

(pudding.cool)
248 points ksampath02 | 1 comments | 01 Apr 25 21:07 UTC | HN request time: 0s | source
Show context
jl6 ◴[02 Apr 25 07:45 UTC] No.43554486[source]▶
What this mostly seems to demonstrate is that hip-pop is endlessly derivative. That might be a consequence of their data source:

> To build this project, we used the dataset of hundreds of thousands of songs on Genius.com accessible through their API, over 200,000 of which were “connected” in some way by sample, interpolation, cover, or remix.

Genres where sampling is openly and explicitly acknowledged are going to be massively over-represented. It would be cool build a relationship network using feature extraction on the actual audio.

replies(4): >>43554923 #>>43556862 #>>43558859 #>>43559230 #
1. Agraillo ◴[02 Apr 25 09:15 UTC] No.43554923[source]▶
I wonder whether a possible audio similarities analyzer would be capable of detecting influences described by the creators themselves. For example, in a podcast, Tal Bachman said that making his most famous hit "She's So High" started when he heard the chorus of "If It Makes You Happy" by Sheryl Crow in a supermarket. After that revelation I now can not unhear it and the choruses of those songs are connected in my brain. But the connection is not that obvious.