i am interested in AI music classification and trying to see if it is possible to learn how to do it.
i have about 30,000 dance tracks and wonder if i can sort them by what i think would appeal to me.
it is a humongous task and recently learned that i am not alone in this pursuit.
when i visit a nightclub and hear music that i love i like to befriend the DJ and ask them how they select music. i learn interesting things like one of my friends associates a mental image of the track's album cover - an older brain function associated with vinyl record covers and now with Soundcloud images (incidentally, the DJ's i know are highly skilled with over 30 years on the job).
classifying music that one likes with Deep Learning methodologies is incredibly hard and unique to every individual but it's worth a try.
here is one take on it ~ if machine learning and data science appeals then take this link:
https://www.datacamp.com/community/tutorials/ML-NLP-lyric-analysis