Home // AIMEDIA 2025, The First International Conference on AI-based Media Innovation // View article
Authors:
Xinyuan Zhu
Clement Leung
Keywords: AI-assisted music creation; Audio fingerprinting; Spectrogram matching; Convolutional neural networks; Audio signal processing
Abstract:
This study explores the application of AI-assisted techniques in analyzing and classifying bowed string instruments from Chinese and Western traditions, focusing on the compar- ison between the erhu and the violin. Using a combination of spectrogram analysis, Mel-Frequency Cepstral Coefficients (MFCCs), and Convolutional Neural Networks (CNNs), the study captures the distinct timbral and articulation differences between the two instruments. Particular attention is given to bowing techniques such as vibrato, portamento, pizzicato, which manifest differently due to structural and acoustic variations. Beyond recognition, this research contributes to AI-assisted music arrangement and composition, providing tools to analyze and synthesize playing techniques across different musical traditions. By bridging Eastern and Western bowed instrument performance styles, this approach supports both cultural heritage preservation and innovation in contemporary music production.
Pages: 7 to 12
Copyright: Copyright (c) IARIA, 2025
Publication date: July 6, 2025
Published in: conference
ISBN: 978-1-68558-330-9
Location: Venice, Italy
Dates: from July 6, 2025 to July 10, 2025