Music Mood Classification of Television Theme Tunes
نویسندگان
چکیده
This paper introduces methods used for Music Mood Classification to assist in the automated tagging of television programme theme tunes for the first time. The methods employed use a knowledge driven approach with tailored parameters extractable from the Matlab MIR Toolbox [1]. Four new features were developed, three based on tonality and one on tempo, to enable a degree of quantified tagging, using support vector machines, employing various kernels, optimised along six mood axes. Using a “nearest neighbour” method of optimisation, a success rate in the range of 80-94% was achieved in being able to classify musical audio on a five point mood scale.
منابع مشابه
Musical Structure in Irish Traditional Tunes
Most Irish traditional folk dance music contains two musical themes; the tune begins with one musical theme which is usually repeated, then progresses to another theme with similar musical structure and motives, which is usually also repeated. The first section is typically called the A section while the second is called the B section. Our goal is to understand, at a quantitative level, the mus...
متن کاملMusical Moods: A Mass Participation Experiment for Affective Classification of Music
In this paper we present our mass participation experiment, Musical Moods. This experiment placed 144 theme tunes online, taken from TV and radio programmes from the last 60 years of the British Broadcasting Corporations (BBC) output. Members of the public were then invited to audition then rate these according to a set of semantic differentials based on the affective categories of evaluation, ...
متن کاملFaceted music: towards a model of music classification
The organization of music is a subject that has fascinated classification researchers and librarians alike for over a hundred years. This paper identifies five key methodological approaches undertaken by commentators on music knowledge organization, which demonstrate different interdependent relationships between musicology and classification. Five significant themes form the main body of this ...
متن کاملCross-cultural Music Mood Classification: A Comparison on English and Chinese Songs
Most existing studies on music mood classification have been focusing on Western music while little research has investigated whether mood categories, audio features, and classification models developed from Western music are applicable to non-Western music. This paper attempts to answer this question through a comparative study on English and Chinese songs. Specifically, a set of Chinese pop s...
متن کاملA Demonstrator for Automatic Music Mood Estimation
Interest in automatic music mood classification is increasing because it could enable people to browse and manage their music collections by means of the music’s emotional expression complementary to the widely used music genres. We continue our work on designing a well defined ground-truth database for music mood classification and show a demonstrator of automatic mood estimation. While a subj...
متن کامل