期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2016
卷号：4
期号：1
页码：827
DOI：10.15680/IJIRCCE.2016.0401051
出版社：S&S Publications
摘要：If a user wants to listen a song and he/she is unable to recognize the title of a song or its related details, then the most direct and convenient method to search for the song is, by humming a section of it, by this approach user will feel more delightful. In the existing system, Q uery by Singing/H umming make use of signal processing or music c omparison methods. Also user's background may get influenced towards the genres of the so ngs being searched. In our study we use the information from the users own database as well as the properties of genres common to users with similar background. To estimate the genre s current user may be interested is based on probability calculat ion, by t his the efficiency of querying by singing/humming is improved. Our system is divided into two phases. The first phase is to search for possible results within music database, while in thesecond phase the musical preference of the user is utilized to rank t he possible results again. Songs that are most likely listen by user would be positioned at the start of the list in the search result. Through our experiments we confirmed that the method proposed can significantly improve the search accuracy
