[DoodleBob]

This is an interesting use of 'machine learning' techniques to decompose petabytes of streamed music data into what makes up successful songs.

What seems likely, however, is that these analytics are being used by record companies to continue feeding the public more of what sells at this time: short, trashy pop songs.

This also means, that there is likely a vast, untapped audience waiting for something good to come along. A fortune awaits the next band that cracks that code.

[DoodleBob]

For your perusal.

[sauropod]

.

[Dr. Sivana]

When Neil Sedaka, who was a serious musician as well as pop artist song-writer, started out, he spent a lot of time analyzing the hits of the late ‘50s, looking for common traits, and churning out more of the same. He was very good at it, and kept writing songs that scored big well into the ‘70s (Bad Blood, Love Will Keep Us Together).

I find the word cloud approach is misleading for all but the youngest demographic. We don’t all listen to Spotify.

[Inyo-Mono]

Music producers and radio executives could study what makes a successful song from a data perspective, helping them more fully understand why people listen to the songs that they do.

As opposed to actually listening to song to hear if is good. That's why radio has gone downhill in the last decade.

[who_would_fardels_bear]

There are some interesting YouTube videos showing that a lot of pop tunes have the same beat and melodies. So much so that they can be played on top of each other and sync up well.

Giving execs more commonality factors between hit songs will only tend to make modern songs even more indistinguishable.

[JohnnyP]

This is a little off topic, but I’ve watched a few “reaction” videos, and have found this “voice coach” fascinating. Here she reacts to Don’t Stop Believing:

https://www.youtube.com/watch?v=kYTPbw_4KfM

Her channel doesn’t have much music I would listen to, but her expressions are fun to watch.

[ShadowAce]

Tech Ping

[Psalm 73]

If you like this topic, check out Mark Steyn’s interview with Randy Bachman - fascinating stuff.

[Lurkina.n.Learnin]

I only have two genres. I like it or I don’t.