The Analysis of Chinese and Japanese Traditional Opera Tunes With Artificial Intelligence Technology Based on Deep Learning

Abstract

This study aims to propose a comparison model of Chinese and Japanese traditional opera tunes based on deep learning (DL) and artificial intelligence technology, which can be used to effectively analyze and classify the tune characteristics of Chinese and Japanese traditional opera. To achieve this aim, a multi-layered DL model, including Convolutional Neural Network, Long Short-Term Memory, Recurrent Neural Network, and Gated Recurrent Unit, is constructed based on the tune characteristics of Chinese and Japanese traditional opera. In the extensive experimental verification, the proposed model has achieved remarkable results in the task of classifying Chinese and Japanese traditional opera tunes. Specifically, the proposed model achieves an accuracy of 89% and 88%, recall of 88% and 87%, and F1-score of 88% and 87% in the tune classification task of the Chinese and Japanese traditional opera, respectively. Compared with similar models, the proposed model performs better in classification.

Keywords

• Chinese and Japanese traditional opera tunes
• deep learning
• artificial intelligence
• tune classification
• convolutional neural network