Emotional Accompaniment Generation System Based on Harmonic Progression

Abstract

A music piece consists of melody and accompaniment in many genres. In this paper, we present a system to automatically generate accompaniment that evokes specific emotions for a given melody. In particular, we propose harmonic progression and onset rate as two key features for emotion-based accompaniment generation. The former refers to the progression of chords, and the latter refers to the number of music events (such as notes and drums) in a unit time. The harmonic progression and the onset rate are altered according to the specified emotion represented by the valence and arousal parameters, respectively. The performance of the system is evaluated subjectively, and the result shows a perfect positive Spearman correlation between the specified emotion and the perceived emotion.

AI-Based Affective Music Generation Systems: A Review of Methods, and Challenges

“…Instead of composing affective music from scratch, hybrid AI-AMG systems are also designed to build the affective music on a pre-composed melody [7,8]. In this case, the pre-composed melody is fed to the system as input along with the emotion-related information.…”

“…Based on the evidence collected from human-based assessment, these systems can generate music to reliably express different emotions from discrete domains and valence-arousal plane. These AI-AMG systems are used for applications such as adaptive music generation for gaming [32], accompaniment generation [8], physiology-sensitive affective music generation [28], etc.…”

Development of an Automatic Emotional Music Accompaniment System by Fuzzy Logic and Adaptive Partition Evolutionary Genetic Algorithm

“…Although these studies provide good composing abilities, the repeatability of the composed phrases is very high and the generated music is independent of emotions. Based on the emotional expression of the harmonic progression, the literature [25] gives a good solution for emotion based composition. This method is processed according to the valence values of chord pairs, and the generated music is indeed able to meet the needs of emotions.…”