CANKO SOCIETY FOR AI AND SOCIAL VALUE

Abstract

In the field of music performance, conducting plays a pivotal role in delivering expressive musical interpretation and ensuring ensemble coordination. However, the evaluation and education of conducting skills have traditionally relied on subjective feedback and limited real-time assessment tools. With the advent of artificial intelligence (AI) and computer vision technologies, it is now possible to analyze conductors’ gestures, tempo control, and expressive intent with greater objectivity and precision. This study aims to develop a personalized education platform that utilizes AI to analyze and assess music conducting performances. The platform employs deep learning techniques, including motion capture, pose estimation, and gesture classification, to extract key features from conductors’ movements. These features are used to evaluate technical proficiency, expressive accuracy, and adherence to musical scores. Based on this analysis, the system provides tailored feedback and learning modules designed to improve individual conducting capabilities. The research methodology involves the collection of conducting video data from both novice and expert conductors, the annotation of key conducting gestures, and the training of neural network models to recognize and evaluate these gestures. Furthermore, the study integrates user-centered design principles to ensure the platform meets educational needs and provides an intuitive interface for students and instructors alike. The expected outcomes of this research include the enhancement of conductor training through objective feedback, the democratization of conducting education by making high-quality instruction more accessible, and the advancement of interdisciplinary applications between music and AI. This study contributes to the innovation of music pedagogy and opens new possibilities for AI-assisted performance training.

Authors

• AI (First Author), Machine – ai.social.value@gmail.com