Human-computer Interaction Based Music Emotion Visualization System and User Experience Assessment

The perceptual and auditory standard of music is deeply integrated with the emerging multimedia to a higher degree, thus forming the music visualization. It is a process presentation method, which provides a brand-new way of interpretation and deduction for music appreciation. In this article, the application of computer aided design (CAD) in music emotion visualization system is studied, and a mapping model between music characteristics and emotion for digital music emotion recognition is constructed by combining with convolutional neural network (CNN). Combined with CAD technology, the structural music features are extracted and calculated, and the main melody and auxiliary melody of music are obtained. Then, based on the separated main melody and auxiliary melody, comprehensive visualization design is carried out to realize the visualization method of highlighting the main melody. In the experimental part, the performance of music emotion recognition algorithm is tested and the user experience is assessed. The results show that the simulation accuracy and user interaction experience of this system have achieved good results, which can improve the interaction between CAD design and viewing of music emotion visualization. Compared with the recurrent neural network (RNN), support vector machine (SVM) and other emotion recognition models, this model has a higher recognition rate of music emotion, which is of great significance to the research of music emotion visualization system. © 2024 U-turn Press LLC.