Music-inspired texture representation.

Horsburgh, Ben; Craw, Susan; Massie, Stewart

Abstract

Techniques for music recommendation are increasingly relying on hybrid representations to retrieve new and exciting music. A key component of these representations is musical content, with texture being the most widely used feature. Current techniques for representing texture however are inspired by speech, not music, therefore music representations are not capturing the correct nature of musical texture. In this paper we investigate two parts of the well-established mel-frequency cepstral coefficients (MFCC) representation: the resolution of mel-frequencies related to the resolution of musical notes; and how best to describe the shape of texture. Through contextualizing these parts, and their relationship to music, a novel music-inspired texture representation is developed. We evaluate this new texture representation by applying it to the task of music recommendation. We use the representation to build three recommendation models, based on current state-of-the-art methods. Our results show that by understanding two key parts of texture representation, it is possible to achieve a significant recommendation improvement. This contribution of a music-inspired texture representation will not only improve content-based representation, but will allow hybrid systems to take advantage of a stronger content component.

Citation

HORSBURGH, B., CRAW, S. and MASSIE, S. 2012. Music-inspired texture representation. In Proceedings of the 26th Association for the Advancement of Artificial Intelligence conference on artificial intelligence (AAAI-12), co-located with the 2012 Symposium on educational advances in artificial intelligence (EAAI-12), 22-26 July 2012, Toronto, Canada. Palo Alto: AAAI Press [online], pages 52-58. Available from: https://www.aaai.org/ocs/index.php/AAAI/AAAI12/paper/view/5041

• Conference Name: 26th Association for the Advancement of Artificial Intelligence conference on artificial intelligence (AAAI-12)
• Conference Location: Toronto, Canada
• Start Date: Jul 22, 2012
• End Date: Jul 26, 2012
• Acceptance Date: Jul 31, 2012
• Online Publication Date: Jul 31, 2012
• Publication Date: Nov 7, 2012
• Deposit Date: Sep 24, 2013
• Publicly Available Date: Sep 24, 2013
• Publisher: Association for the Advancement of Artificial Intelligence
• Pages: 52-58
• ISBN: 9781577355687
• Public URL: http://hdl.handle.net/10059/871
• Publisher URL: https://www.aaai.org/ocs/index.php/AAAI/AAAI12/paper/view/5041