Evaluating the Impact of Sound Events’ Rhythm Characteristics to Listener’s Valence
JOURNAL OF THE AUDIO ENGINEERING SOCIETY(2015)
摘要
Sound is a prominent element in the communication of humans with their environment. It can be either in an organized form, e.g., speech or music, or in a non organized form, i.e., Sound Events (SEs). There are many studies focusing on emotion recognition from music content, but the research on emotion recognition from SEs is a rather new field. This works falls within this field and concludes that the rhythmic characteristics of an SE have a major impact on the listener's arousal. This result verifies the empirical knowledge that the rhythm of sound affects the listener's activation state. Moreover, we also investigate whether the above characteristics also affect the pleasure of the listener Toward this aim, we have utilized a well known data set of emotionally annotated SEs, extracted various rhythm-related technical cues, and conducted a series of machine learning experiments. The overall results indicate a relation of the rhythm and listener's valence with accuracy results reaching up to 63%.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络