Comparison and Analysis of Acoustic Features of Western and Chinese Classical Music Emotion Recognition Based on V-A Model

Music emotion recognition is increasingly becoming important in scientific research and practical applications. Due to the differences musical characteristics between Western Chinese classical music, it necessary investigate distinctions music emotional feature sets improve accuracy of cross-cultural models. Therefore, a comparative study on was conducted. Using V-A model as an perception model, approximately 1000 pieces excerpts total were selected, 20-dimension for different dimensions datasets finally extracted. We considered kinds algorithms at each step training process, from pre-processing selection regression selection. The results reveal that combination MaxAbsScaler wrapper method using recursive elimination algorithm based extremely randomized trees optimal algorithm. harmonic change detection function culturally universal feature, whereas spectral flux specific music. It also found pitch features are more significant loudness rhythm