Extracting Emotions from Music Data Alicja Wieczorkowska 1 , Piotr Synak 1 , Rory Lewis 2 , and Zbigniew Ras 2 1 Polish-Japanese Institute of Information Technology, Koszykowa 86, 02-008 Warsaw, Poland, alicja,synak@pjwstk.edu.pl, 2 University of North Carolina, Charlotte, Computer Science Dept., 9201 University City Blvd., Charlotte, NC 28223, USA, rorlewis,ras@uncc.edu Abstract. Music is not only a set of sounds, it evokes emotions, subjec- tively perceived by listeners. The growing amount of audio data available on CDs and in the Internet wakes up a need for content-based searching through these files. The user may be interested in finding pieces in a spe- cific mood. The goal of this paper is to elaborate tools for such a search. A method for the appropriate objective description (parameterization) of audio files is proposed, and experiments on a set of music pieces are described. The results are summarized in concluding chapter. 1 Introduction Extracting information on emotions from music is difficult for many reasons. First of all, music itself is a subjective quality, related to culture. Music can be defined in various ways, for instance, as an artistic form of auditory communi- cation incorporating instrumental or vocal tones in a structured and continuous manner [31], or as the art of combining sounds of voices or instruments to achieve beauty of form and expression of emotion [6]. Therefore, music is inseparably related to emotions. Musical structures itself communicate emotions, and also synthesized music aims at expressive performance [13], [19]. The experience of music listening can be considered within three levels of human emotion [12]: – autonomic level, – denotative (connotative) level, and – interpretive (critical) level. According to [17], music is heard: – as sound. The constant monitoring of auditory stimuli does not switch off when people listen to music; like any other stimulus in the auditory environ- ment, music is monitored and analyzed. – as human utterance. Humans have an ability to communicate and detect emotion in the contours and timbres of vocal utterances; a musical listening experience does not annihilate this ability.