Disambiguating Music Emotion Using Software Agents

Annotating music poses a cognitive load on listeners and this potentially interferes with the emotions being reported. One solution is to let software agents learn to make the annotator’s task easier and more efficient. Emo is a music annotation prototype that combines inputs from both human and software agents to better study human listening. A compositional theory of musical meaning provides the overall heuristics for the annotation process, with the listener drawing upon different influences such as acoustics, lyrics and cultural metadata to focus on a specific musical mood. Software agents track the way these choices are made from the influences available. A functional theory of human emotion provides the basis for introducing necessary bias into the machine learning agents. Conflicting positive and negative emotions can be separated on the basis of their different function (reward-approach and threat-avoidance) or dysfunction (psychotic). Negative emotions have strong ambiguity and these are the focus of the experiment. The results of mining psychological features of lyrics are promising, recognisable in terms of common sense ideas of emotion and in terms of accuracy. Further ideas for deploying agents in this model of music annotation are presented.

[1]  Marc Leman,et al.  Correlation of Gestural Musical Audio Cues and Perceived Expressive Qualities , 2003, Gesture Workshop.

[2]  François Pachet,et al.  Evolving Automatically High-Level Music Descriptors from Acoustic Signals , 2003, CMMR.

[3]  Stephan Baumann,et al.  Super-convenience for Non-musicans: Querying MP3 and the Semantic Web , 2002, ISMIR.

[4]  Page Section 7 , 1988 .

[5]  Patrick Brézillon,et al.  Lecture Notes in Artificial Intelligence , 1999 .

[6]  Dave Datta Managing Metadata , 2002, ISMIR.

[7]  J. Russell Core affect and the psychological construction of emotion. , 2003, Psychological review.

[8]  Barry Vercoe,et al.  Using User Models in Music Information Retrieval Systems , 2000, ISMIR.

[9]  K. Scherer Toward a dynamic theory of emotion : The component process model of affective states , 1987 .

[10]  David Watson,et al.  The PANAS-X manual for the positive and negative affect schedule , 1994 .

[11]  George Tzanetakis,et al.  Manipulation, analysis and retrieval systems for audio signals , 2002 .

[12]  J. Cacioppo,et al.  The affect system has parallel and integrative processing components: Form follows function. , 1999 .

[13]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques with Java implementations , 2002, SGMD.

[14]  Marshall S. Smith,et al.  The general inquirer: A computer approach to content analysis. , 1967 .

[15]  D. Watson Mood and temperament , 2000 .

[16]  Lie Lu,et al.  Automatic mood detection from acoustic music data , 2003, ISMIR.

[17]  J. Stephen Downie,et al.  Toward the scientific evaluation of music information retrieval systems , 2003, ISMIR.

[18]  D. Watson,et al.  On the Dimensional and Hierarchical Structure of Affect , 1999 .

[19]  J. Kenney Mathematics of statistics , 1940 .

[20]  G. A. Mendelsohn,et al.  Affect grid : A single-item scale of pleasure and arousal , 1989 .