In this paper we introduce a new key finding algorithm that is based on the symmetry model introduced by Gatzsche et al. in [1], [2], [11]. The algorithm consists of two parts: First the most probable diatonic pitch class set of the musical piece is recognized. Second using one of the subspaces of the symmetry model the mode of the piece is estimated. The algorithm is evaluated with 100 Beatles songs, 90 newer Pop&Rock songs and 252 classical pieces from the Naxos database. The results will be compared to the algorithms of Lerch [3], Zhu et al. [4] and an algorithm based on binary major and minor chord profiles. The new algorithm has the highest overall key finding MIREX’05 score of 82,9 %.

Authors: Mehnert, Markus; Gatzsche, Gabriel; Arndt, Daniel
Affiliations: Technische Universität Ilmenau, Ilmenau, Germany; Fraunhofer Institute for Digital Media Technology, Ilmenau, Germany; Fraunhofer IIS, Ilmenau, Germany(See document for exact affiliation information.)
AES Convention: 126 (May 2009) Paper Number: 7660
Publication Date: May 1, 2009 Import into BibTeX
Subject: Audio for Games and Interactive Media
Permalink: https://www.aes.org/e-lib/browse.cfm?elib=14856

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 and would like to subscribe to the E-Library then Join the AES!

This paper costs $33 for non-members and is free for AES members and E-Library subscribers.

Learn more about the AES E-Library

E-Library Location: (CD 126Papers)   /126/7660.pdf

Start a discussion about this paper!