In the age of digital recordings, many institutions maintain large databases of concert recordings. While segmentation of these concert recordings for mastering and production is a time-consuming task for humans, this paper presents a novel heuristic algorithm to automate that process. Building on other work in audio segmentation, technique from the music informatics community is used to detect events, classify them, and segment entire concert recordings unsupervised. A brief review of previous work and the methodology used in this approach are provided, as well as the results obtained on a corpus of sixteen concerts.
(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.
