Skip to main content

Topic: music categorization methodologies and implementation (Read 1436 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • groovebug
  • [*]
music categorization methodologies and implementation
I am a newbie to MP3s and this board. I have many questions but am slowly going through all the old post to avoid asking something stupid.

Does anyone have/know of an advanced methodology to categorize music? I have approx 14,000 songs and get another 100 per day thru stream recording. Categorizing by “genre” gives me 3 folders: Rock-Indie-Classic Rock with about 4,000 songs each – this is not good enough. Outside of house music, BPM is of little value… a 115 may “sound” like a slow song, a 090 may sound like a violent, dark rock song.

I’m using MP3 Observer (MP3 filing SW) which allows for entry for fields of: genre-preference-tempo-situation-mood. That’s all well and good, but is there a consistent method for applying these fields? I like what allmusic.com is trying to do, but it is not workable for me: They have 200+ “situations” and 200+ “moods” (what is the standard to determine a “bright” song from an “ambitious” song?) and I don’t agree with their conclusions.

1. Can anyone recommend filing software any better than MP3 Observer?
2. Does anyone know of a published, consistent, reasoned approach (as least subjective as possible) for applying aesthetic fields? I’m imagining a workable system would have no more that 5 variables per field.
3. Is there software that will analyze the song and populate these fields for me? Is there any way to avoid having to manually tag each field for each song?  (and it will have to be “per song”… “per artist” will not work for me… see P.Floyd “One of These Days” vs. “Us and Them”)

Any advice/comments would be most helpful to me.