Many of today’s popular dance music records are identifiable by a ”drop”—a section of the song that is commonly the highest in both listener-perceived and actual signal energy. In this paper we examine several computational methods for locating the exact time at which the drop occurs in a given audio sample. Various metrics are compared and contrasted based on relevant audio signal features. This technology has potential applications within automated DJ software, online music streaming services, computational ethnomusicology research, and more.
(See document for exact affiliation information.)
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.
