Designing a Deep Learning-Enabled Music Teaching System in Universities Using the Moodle Platform

Abstract

Music education plays a vital role in fostering creativity, expression, and cognitive development among students in university settings. Moodle is the learning management platform to promotes significant knowledge sharing among the students in the Universities. In this paper, introduce the Federated Deep Learning Moodle Hidden Chain (FDLMHc) with the Moodle platform for music education experiences. The FDLMHc system combines the power of federated learning with the flexibility of Moodle to provide personalized feedback and adaptive learning pathways for students. The FDLMHc model uses the Music signal pitch estimation with the consideration of the different pitches in the signal frequencies. The signal of the music signal is estimated for the different SNR rates of -10dB, 0dB, 10 dB, and 20 dB. The proposed FDLMHc model computes and processes the music signal with the hidden chain process for the estimation of the pitches in the music signal. The estimated hidden chain model is applied over the federated learning network for the classification of the signal in the Music. The findings reveal promising results, demonstrating the system's ability to accurately classify musical elements, such as pitch, rhythm, and dynamics, while providing personalized feedback tailored to individual student needs. The accuracy for the estimation of the Music pitch is estimated as the 95% with a convergence rate of 91% for the estimation of the signal in the Music signal.