Classical Music Specific Mood Automatic Recognition Model Proposal

Suyeon Lee,Haemin Jeong,Hyeyoung Ko

doi:10.3390/electronics10202489

Abstract

The purpose of this study was to propose an effective model for recognizing the detailed mood of classical music. First, in this study, the subject classical music was segmented via MFCC analysis by tone, which is one of the acoustic features. Short segments of 5 s or under, which are not easy to use in mood recognition or service, were merged with the preceding or rear segment using an algorithm. In addition, 18 adjective classes that can be used as representative moods of classical music were defined. Finally, after analyzing 19 kinds of acoustic features of classical music segments using XGBoost, a model was proposed that can automatically recognize the music mood through learning. The XGBoost algorithm that is proposed in this study, which uses the automatic music segmentation method according to the characteristics of tone and mood using acoustic features, was evaluated and shown to improve the performance of mood recognition. The result of this study will be used as a basis for the production of an affect convergence platform service where the mood is fused with similar visual media when listening to classical music by recognizing the mood of the detailed section.

Highlights

Music of various moods has been shown to enrich the listener’s emotions and promote psychological and mental stability [1,2,3,4,5]
In order to improve the performance of recognizing the detailed mood flow of classical music, in the previous chapter, the music segment extraction method, which is the unit of mood recognition, was applied and a model capable of classifying 18 moods was produced
The receiver operating characteristic (ROC) is an index showing a pair of the percentage of samples predicted to be positive among all positive samples and the percentage of samples that were incorrectly predicted to be positive among all negative samples as a curve

Summary

Introduction

Music of various moods has been shown to enrich the listener’s emotions and promote psychological and mental stability [1,2,3,4,5]. Many studies are being performed in the field of music emotion recognition (MER) to classify or recommend music based on emotion. In the case of music with lyrics, there is a study addressing the recognition of emotions by interpreting the lyrics [11,12]. As there are cases in which there are no lyrics or the structural characteristics of music are more related to emotions rather than the lyrics, many studies are being conducted to recognize the emotions of music by analyzing the acoustic features of the music. Various studies related to the service application fields, tailored to the sensibility of music, are being conducted [13,14,15,16]

Objectives

Methods

Results

Discussion

Conclusion