Home // BRAININFO 2025, The Tenth International Conference on Neuroscience and Cognitive Brain Information // View article

Dynamic Emotion Analysis in Piano Music Based on Performance Techniques Recognition

Authors:
Yueyan Wu
Clement Leung

Keywords: Performance Techniques Recognition; Convolutional Neural Network (CNN); Music Emotion Recognition (MER).

Abstract:
The relationship between music and emotion has always been essential in musicology and psychology. This study aims to automatically identify the playing technique in piano performance through deep learning technology and analyze its influence on the dynamic change of emotion. We propose a technique recognition method based on a deep Convolutional Neural Network (CNN), which can accurately identify different techniques (such as octave, vibrato, glissando, etc.). In addition, we design a simple temporal analysis model to analyze the evolution of emotion over time based on the dynamic change of playing technique. The experimental results show that the identification of playing techniques achieves nearly 86% accuracy, outperforming traditional methods, and specific playing techniques are significantly related to certain emotions. There are also results on the dynamic emotion analysis task. This study not only provides a new perspective and method for the field of music emotion recognition but also provides a new tool and method for music analysis and music education.

Pages: 1 to 6

Copyright: Copyright (c) IARIA, 2025

Publication date: March 9, 2025

Published in: conference

ISSN: 2519-8653

ISBN: 978-1-68558-239-5

Location: Lisbon, Portugal

Dates: from March 9, 2025 to March 13, 2025