Machine Recognition and Subconscious Perception Characteristics of Musical Emotions

Music, an auditory art form abundant in emotional characteristics, plays a crucial role in human sensory experiences and psychological adjustments. Its emotional impact originates from the brain's emotional mechanisms, music appreciation education, as well as the surrounding environment and mental state. These effects involve subconscious perception and a complex interaction of emotions. In recent years, significant progress has been made in machine recognition of musical emotions, driven by the rapid development of artificial intelligence technologies. This paper presents a systematic review and analysis of relevant research, concentrating on aspects such as emotional feature extraction from music, emotional model construction, machine recognition methods, and the subtle perceptual characteristics of music. The paper proposes four classifications for music emotional features: basic acoustic features, music perception features, neural activity features, and multi-dimensional fusion features. Additionally, it summarizes four machine recognition methods based on audio data, lyrics, neural signals, and multimodal data. The discussion also addresses current challenges and outlines future research directions, aiming to provide insights for the advancement of music emotion research and intelligent recommendation technologies.