Innovation of teaching mechanism of music course integrating artificial intelligence technology: ITMMCAI-MCA-ACNN approach

Abstract

The major objective is to teach that makes use of interactive, intelligent technologies, as well as customized utilizing examples to examine concepts in theory and the development of practical skills. The manuscript introduced a music teaching system called Attention-based Convolutional Neural Network (ITMMCAI-MCA-ACNN). The system uses data from the Musdb18 dataset and a pre-processing step is performed to remove noise and imperfect records using the Horizontal Gradient Filter. Subsequently, the pre-processed data is passed through a source separationusing Attention-based convolutional neural network (ACNN)optimized withMusical chairs optimization approach to isolate different audio components like drums, bass, vocals, and other sounds, from a mixed audio signal for effective music teaching. The proposed ITMMCAI-MCA-ACNN is implemented in MATLAB, using the Musdb18 dataset for evaluation examination. The proposed method’s efficacy is measured using several success indicators, including precision, accuracy, specificity, error rate, sensitivity, and F1-score. The effectiveness of the suggested ITMMCAI-MCA-ACNN technique works 75.89 %, 61.11 %, and86%high accuracy, and90%, 73 %, and 70 % high precision compared with existing methods such as ITMMCAI-AIT, ITMMCAI-AIT-WN, and ITMMCAI-MDCT respectively.