Intelligent generation and optimization of resources in music teaching reform based on artificial intelligence and deep learning

Abstract

In order to increase the effectiveness and personalization of music instruction, this paper aims to create a deep reinforcement learning (DRL)-based framework for creating music resources. Therefore, a Melody Generation Model in Music Education Based on Actor-Critic Framework (AC-MGME) is proposed. This model analyzes students’ learning status in real time through AC-MGME algorithm, generates melodies that match their abilities, and enhances the polyphonic generation effect by using multi-label classification and attention mechanism. According to the testing results, the proposed model clearly outperforms the baseline Deep Q-Network (DQN) algorithm, achieving 95.95% accuracy and 91.02% F1 score in melody generation quality with a generation time of 2.69 s. Therefore, the constructed model can not only generate high-quality personalized melody, but also shows a significant improvement in improving user experience and learning effect, providing reference direction for the generation and optimization of intelligent resources in music teaching.