Event Abstract Back to Event Exploring relationships between audio features and emotion in music Cyril Laurier1*, Olivier Lartillot2, Tuomas Eerola2 and Petri Toiviaine2 1 Music Technology Group, Universitat Pompeu Fabra, Spain 2 Department of Music, University of Jyväskylä , Finland Background: Detecting emotions in music from audio content is a relatively recent problem. Only a few works investigate the relationship between features extracted from audio and emotion dimensions or categories. Aims: The goal is to analyze the associations between emotion categories, dimensions, and audio features automatically extracted from audio content. Method: Our research is based on ground truth data created in a study by Eerola & Vuoskoski (submitted), where 110 15-second excerpts from film soundtracks were evaluated by 116 participants using 5 target emotions (fear, anger, happiness, sadness, tenderness) and 3 dimensions (valence, energy arousal and tension arousal). Exploiting state-of-the-art techniques, we extract audio features and study the correlations between them and the annotations. We also investigate which descriptors are the most relevant to classify music by emotion. Results: First, we observe that the energy dimension is the most informative to differentiate between emotion categories. Then, we report on the relationships between audio descriptors and annotated categories and dimensions. We notice a correlation of mode with valence. We also highlight the relevancy of spectral descriptors. Conclusions: Based on music made to convey emotions (soundtracks), we confirm results in psychology, and we open the discussion around other relevant audio cues. Conference: Tuning the Brain for Music, Helsinki, Finland, 5 Feb - 6 Feb, 2009. Presentation Type: Poster Presentation Topic: Posters Citation: Laurier C, Lartillot O, Eerola T and Toiviaine P (2009). Exploring relationships between audio features and emotion in music. Conference Abstract: Tuning the Brain for Music. doi: 10.3389/conf.neuro.09.2009.02.033 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 26 Jan 2009; Published Online: 26 Jan 2009. * Correspondence: Cyril Laurier, Music Technology Group, Universitat Pompeu Fabra, Barcelona, Spain, cyril.laurier@upf.edu Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Cyril Laurier Olivier Lartillot Tuomas Eerola Petri Toiviaine Google Cyril Laurier Olivier Lartillot Tuomas Eerola Petri Toiviaine Google Scholar Cyril Laurier Olivier Lartillot Tuomas Eerola Petri Toiviaine PubMed Cyril Laurier Olivier Lartillot Tuomas Eerola Petri Toiviaine Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.