The separation of music into genres based on the various audio characteristics of songs is a classification task performed both by those who study music and those who listen to and enjoy music. Songs that share similar roots, sound, and instruments are generally grouped together into a genre. Today, large music-streaming services such as Spotify have millions of songs available in their online catalogs, making it impractical to manually classify by genre. Neural networks, a form of artificial intelligence inspired by the learning processes of the human brain, excel at finding patterns and organizing data into groups based on features of the data. This research project created and refined a neural network that can identify a song’s genre based on specific audio features.
Doolittle, Austin, "GeNNre: Automated Music Genre Classification" (2017). Honors Projects. 634.