Open Access Article

Title: Sparse coding-based vocal music feature extraction and real-time transmission

Authors: Fangzi Zhang; Jinyi Hu

Addresses: School of Music, Hunan International Economics University, Changsha, 410000, China ' School of Humanities and Music, Hunan Vocational College of Science and Technology, Changsha, 410000, China

Abstract: Traditional audio compression and transmission methods struggle with bandwidth usage and transmission delay, thereby creating a growing need for a real-time audio transmission. This work presents a sparse coding-based approach for vocal audio feature extraction and real-time transmission (SCTRT) to handle these difficulties. By means of sparse coding approaches, the model efficiently compresses and extracts audio information, hence lowering data redundancy and improving transmission efficiency. Three components make up the model: real-time transmission and recovery, feature extraction and compression, and audio capture and pre-processing, guaranteeing low latency and effective transmission of audio signals. In terms of compression ratio, audio quality and transmission delay, the experimental findings reveal that the SCTRT model is particularly appropriate for real-time audio transmission applications since it has notable benefits over conventional techniques.

Keywords: sparse coding; vocal feature extraction; audio compression; real-time transmission.

DOI: 10.1504/IJICT.2025.150145

International Journal of Information and Communication Technology, 2025 Vol.26 No.42, pp.1 - 17

Received: 30 Dec 2024
Accepted: 14 Jan 2025

Published online: 01 Dec 2025 *