Font Size: a A A

Research On Music Emotion Recognition Model

Posted on:2021-01-06Degree:MasterType:Thesis
Country:ChinaCandidate:J J YanFull Text:PDF
GTID:2415330611989286Subject:Detection Technology and Automation
Abstract/Summary:PDF Full Text Request
As an important part of human life,music can convey emotion and regulate the mood of the audience.emotion is the essential characteristic of music,and the relationship between music and emotion has become the subject of many academic discussions and empirical studies,involving many different disciplines,including philosophy,musicology,psychology,biology,anthropology and sociology.At present,with the rapid development of information technology and artificial intelligence,the research of music emotion recognition is progressing rapidly,which has become one of the important research directions in the field of digital music.This paper studies the construction and application of music emotion recognition model,the main research work is as follows:(1)Research on music feature analysis technology: by analyzing the structure composition of MIDI files,using contour line algorithm based on audio feature improvement to identify the main melody of music,and on this basis,extract the emotional features of music and carry out standardized processing,as the input of classification model.(2)Establishment of music emotion recognition model: the emotional psychology model is established according to the data scale,and four basic emotions are used as the result of music emotion classification.The multi-classification model of emotion is built by using the Support Vector Machine(SVM)classifier,and the parameters of the model are optimized by Particle Swarm Optimization(PSO)algorithm.(3)This paper collects 231 music fragments and carries on the manual annotation to train the classifier.In addition,this paper improves the classification model on the basis of main melody extraction and music emotion feature analysis,and applies the recognition model to stage lighting control,and obtains good simulation results.Through the experimental comparison and the application results of stage lightingcontrol,it shows that the established recognition model has good performance in classification accuracy and test time,which can provide an effective reference for stage lighting design.
Keywords/Search Tags:emotion classification model, music feature extraction, stage lighting
PDF Full Text Request
Related items