Exploring music with a probabilistic projection interface

Vad, B., Williamson, J. , Boland, D., Murray-Smith, R. and Steffensen, P. (2015) Exploring music with a probabilistic projection interface. In: Machine Learning for Music Discovery Workshop, Lille, France, 11 Jul 2015,

Abstract

We present the design and evaluation of an in- teractive tool for music exploration, with musi- cal mood and genre inferred directly from tracks. It uses probabilistic representations of multivari- able predictions of subjective characteristics of the music to give users subtle, nuanced visuali- sations of the 2D map. These explicitly repre- sent the uncertainty and overlap among features and support music exploration and casual playlist generation. A longitudinal trial in users’ homes showed that probabilistic highlighting of subjec- tive features led to more focused exploration in mouse activity logs, and 6 of 8 users preferred the probabilistic highlighting.

• Item Type: Conference Proceedings
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Murray-Smith, Professor Roderick and Boland, Mr Daniel and Williamson, Dr John
• Authors: Vad, B., Williamson, J., Boland, D., Murray-Smith, R., and Steffensen, P.
• College/School: College of Science and Engineering > School of Computing Science
• Copyright Holders: Copyright © 2015 The Authors
• Publisher Policy: Reproduced under a Creative Commons License

Related URLs

• Organisation