{"title":"Music Impression Extraction Method By chord Impressions and Its Application to Music Media Retrieval","authors":"Hiroki Nakata, T. Nakanishi","doi":"10.1109/SNPD51163.2021.9704990","DOIUrl":null,"url":null,"abstract":"In this paper, we represent an impression extraction method for music by relationship between chords and impression terms. Our methods extract impression term with weights from chord extracted from music as wav file. In addition, we created a module for further application of the music impression extraction method. We designed the system which enable us to retrieve music from the word which has been input by user. We use a data set consisting of a music file and annotation file with an impression terms to each song in order to create models that relates to chord extracted from music with an impression using cosine-similarity. By using this first model, we can retrieve music based on the words that we put. We will realize a recommendation system based on the chords from music and our impression of the word.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SNPD51163.2021.9704990","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
In this paper, we represent an impression extraction method for music by relationship between chords and impression terms. Our methods extract impression term with weights from chord extracted from music as wav file. In addition, we created a module for further application of the music impression extraction method. We designed the system which enable us to retrieve music from the word which has been input by user. We use a data set consisting of a music file and annotation file with an impression terms to each song in order to create models that relates to chord extracted from music with an impression using cosine-similarity. By using this first model, we can retrieve music based on the words that we put. We will realize a recommendation system based on the chords from music and our impression of the word.