The Design And Implementation Of Music Comparison Analysis System Based On Melody Similarity

With the increasing development and integration of the music industry and Internet information technology,the music market is rapidly expanding its capacity,which also accompanied by an increasing number of music infringement incidents.The protection of music intellectual property rights has also attracted more and more attention from the government and the legal profession.But now,there is no specific quantitative criteria for the court to determine whether music infringes or not.The judges rely more on their own cognition of music infringement and subjective feelings of music works to judge.Therefore,it is necessary to provide comprehensive and objective music comparison analysis data to assist in the judgment of music infringement cases.In addition,scenes such as the music works rechecking of music majors,the music quality identification of AI composition results,and the daily recognition and comparison of music lovers also need the support of objective data of music comparison analysis system.This thesis designs and implements a music comparison analysis system,which has the functions of music uploading,music comparing,result downloading,case viewing and so on.Among them,music comparing mainly analyzes and compares the melody interval and rhythm of two music parts,gives the detailed similarity and provides objective,understandable and comprehensive music comparison analysis data.People upload original works and suspected similar works,the music comparison analysis system with a constructed music comparison model will display the results of music comparison in multiple dimensions,including overall similarity percentage,the comparison of mode and tonality,top3 of similar segments,specific similar segment position and duration,etc.In the construction of the music comparison model,we finally choose the improved Sunday algorithm as the matching algorithm in the final model by comparing a variety of string matching algorithms,which can further improve the efficiency of music comparison.The music comparison analysis system in this thesis adopts Vue framework and Spring Boot technology to complete the implementation of front-end web page and back-end framework,and stores data information through My SQL database.Through experimental analysis,the efficiency of this comparison algorithm is significantly improved,compared with other matching algorithms,time can be shortened by 40% to95%.In addition,the results of questionnaire show that the comparison result of music comparison analysis in this thesis is more readable.After a series of functional tests and performance tests,the system has realized functional requirements and non-functional requirements,in line with the expected assumption.