An automated method for classifying music signals into genres is proposed. Up to three genres are assigned to each example along with a measure of the degree of influence. The method is based on a structure formed by taxonomy of four hierarchical layers, with 29 genres in the lowest layer and 10 target genres in the higher layers. In more than 77% of the cases there was successful classification. The computational effort is sufficiently low that this method could be used in real time.
(See document for exact affiliation information.)
https://aes2.org/publications/elibrary-page/?id=14400
(336KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
