Traditional music AI technologies mostly focus on understanding audio signals, whereas the highlight of this research is its innovative approach to symbolic music understanding. The paper introduces N-Gram theory into the field of music understanding and proposes a unique unsupervised compound word technique (UCW) to deepen the understanding of symbolic music. At the model architecture level, it proposes the NG-Midiformer model, which incorporates the features of the N-Gram Transformer encoder and greatly enhances the ability of the model to recognize and process musical symbols.

Figure Figure 1 Relationships and differences among REMI, CP and UCW

By applying deep learning and N-Gram techniques, NG-Midiformer can analyze musical sequences in greater depth and effectively capture the subtle connections between musical elements. In addition, this research introduces a feature injection method that incorporates additional musical feature information into the model. These features can include information on aspects such as musical style and emotion, which helps improve the understanding and generation of music. By using N-Gram unsupervised compoundation and feature injection, this research successfully solves several problems in symbolic music understanding and generation. These methods can improve the understanding of musical structure and features, and thus better generate music that matches specific styles and emotions. The model performs excellently on multiple public datasets, achieving industry-leading results.

Figure 2 Overall architecture of the N-gram Transformer encoder

The paper provides a new research perspective and methods for the field of symbolic music understanding. By combining natural language processing technology with the special needs of the music domain, it creates new means for the application of AI to music understanding. This research was supported by the National Natural Science Foundation of China, the Natural Science Foundation of Hubei Province, the Fundamental Research Funds for the Central Universities and other projects.