Abstract:Music is a powerful language capable of eliciting a variety of emotions in individuals. Understanding and recognizing these emotions is pivotal for applications ranging from personalized music recommendations and music therapy to automatic music composition and affective computing. Presently, deep learning for music emotion recognition is gaining popularity, primarily relying on timbre features to capture local spatial information. However, there is an untapped potential in incorporating other pertinent audio features and global correlations in the feature space to capture the repetitive temporal information of music for emotion classification. This study introduces GlocalEmoNet as a method to capture both local and global correlations in music, utilizing timbre and Chroma audio features for tasks related to emotion classification and segmentation. The neural network underwent training and testing on approximately six thousand music audio samples, encompassing six music-emotion categories. The utilization of a genetic algorithm is employed for optimizing the hyperparameters of the proposed neural networks, aiming to attain optimal performance, efficiency, and generalization. The best classifier demonstrated superior performance, surpassing previously published results by a significant margin of approximately 14%. The optimal classifier achieved an accuracy score of 81.66%, an f1-score of 0.812, and an area under the curve score of 0.956. The evaluation of classification and segmentation outcomes also involved the use of visual representations.

Music emotion recognition system based on improved GA-BP

Web Music Emotion Recognition Based on Higher Effective Gene Expression Programming

A New Fuzzy Classifier For Music Emotion Based On Conditional Probability

Using Psychophysiologicalmeasures to Recognize Personalmusic Emotional Experience

Human-centric Music Medical Therapy Exploration System

Gait Phase Recognition of Lower Limb Based on GA Optimized BP Neural Network

Research on Emotional Vocabulary-Driven Personalized Music Retrieval

Musical perceptual similarity estimation using interactive genetic algorithm

Music emotion recognition based on temporal convolutional attention network using EEG

Emotion recognition from sound stimuli based on back-propagation neural networks and electroencephalograms

Computer Analysis and Automatic Recognition Technology of Music Emotion

Emotional Recognition For Chime Bell Music

FFA-BiGRU: Attention-Based Spatial-Temporal Feature Extraction Model for Music Emotion Classification

Emotional Music Generation Using Interactive Genetic Algorithm

Digital Empirical Research of Influencing Factors of Musical Emotion Classification Based on Pleasure-Arousal Musical Emotion Fuzzy Model

EmotionBox: a music-element-driven emotional music generation system using Recurrent Neural Network

User-Adaptive Music Emotion Recognition

Analysis and recognition of a novel experimental paradigm for musical emotion brain-computer interface

A survey of music emotion recognition

GlocalEmoNet: An optimized neural network for music emotion classification and segmentation using timbre and chroma features

A New Multilabel System for Automatic Music Emotion Recognition