Spotify for Classical Music

[rys928]

I use Spotify and something that has always annoyed me is the lack of consistency in the way classical music is organized on the platform. To put it simply, major music services are not built for the classical listener.
Luckily, some companies (Apple, and Idagio) have sought to alleviate the pain in better or worse ways. For those of us who continue to use Spotify, there isn't any way to get around this.

For a class, I exported random tracks from Spotify and ran them through data-cleaning processes to standardize the spelling and format for composers and track names. For those of you not familiar with Spotify URIs, they are specific identifying IDs for tracks and albums. You can use the URI to go straight to a specific track by using the search bar (for example search spotify:track:78oFMCjLNsIMG4n6iYyucj , it will bring you a Beethoven Sonata). Some of the code has to be cleaned up.

When applied to a large scale, something of this nature could make using Spotify more friendly to the Classical listener if built into their platform on a larger scale. It would help keep data clean. (For example, track names will no longer vary and have tracks listed as "Beethoven Symphony No. 3...", "Symphonie No.3...", "Eroica, Symphony No. 3...", or "Klemperer conducts Symphony No.3...". In addition, composers and performers would not share the same "artist" label and be split. The result would be that every track on the platform would have a standardized page for the composer, performers, and track names for each piece.

Let me know what you think about these ideas, based on the small sample provided!

https://drive.google.com/file/d/109dAsl3jooL54IoRcrAbOOV3CDsO9U_Z/view?usp=sharing

[stringoverstrung]

Probably AI will be there before you can complete your project. It will be solved in the relatively near future.

R,
G.

[kosulin]

Most commercially distributed recorded music has ISRC assigned, but so far it did help much to cleanup the mess. I work for a huge music company, and can tell you firsthand - creator and artist names, tracks, releases are nightmare to deal with, especially legacy ones (and most of the data is legacy). A lot of duplicate names and titles, or different names or spellings for the same item, etc.