Skip to main navigation Skip to search Skip to main content

MusicNet: A Real-Time Web Platform for Multiplayer Piano Learning over Unstable Networks

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper presents MusicNet, a browser-based multiplayer platform designed for real-time piano training over the Internet. The system integrates gamification, real-time pitch detection, and an AI-based model for packet loss concealment (PAR-Cnet). Through WebRTC peer-to-peer communication, MusicNet enables users to play, receive visual feedback, and synchronize gameplay across diverse network variations. Our solution ensures interactivity and musical precision by implementing adaptive buffering, packet reconstruction, and latency handling. Results demonstrate improved usability, stability, and learning experience across real-time conditions, with high acceptance observed among users from varied age groups and musical backgrounds during multi-session evaluations.

Original languageEnglish
Title of host publication2025 IEEE 6th International Symposium on the Internet of Sounds, IS2 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages5
ISBN (Electronic)9798331572945
DOIs
StatePublished - 29 Oct 2025
Event6th IEEE International Symposium on the Internet of Sounds, IS2 2025 - L�Aquila, Italy
Duration: 29 Oct 202531 Oct 2025

Publication series

Name2025 IEEE 6th International Symposium on the Internet of Sounds, IS2 2025

Conference

Conference6th IEEE International Symposium on the Internet of Sounds, IS2 2025
Country/TerritoryItaly
CityL�Aquila
Period29/10/2531/10/25

Keywords

  • AI in audio
  • Real-time systems
  • WebRTC
  • multiplayer applications
  • music education
  • packet loss concealment

Fingerprint

Dive into the research topics of 'MusicNet: A Real-Time Web Platform for Multiplayer Piano Learning over Unstable Networks'. Together they form a unique fingerprint.

Cite this