Memory usage for storing indexes and query response times for retrieval processing are two critical issues in music information retrieval (MIR) systems. In this paper, we propose an effective and efficient numeric indexing structure to overcome the difficulties of variable length queries and enhance the efficiency of music retrieval. The proposed structure differs greatly from pre-existing research in textual indexing techniques such asn-gram and suffix tree because it does not need to generate redundant and useless indexes. The index construction process has no complicated split and joint operations making, is easier and faster than tree-like methods. Experimental results show that our method is more scalable and economical than previous methods. The proposed method can significantly reduce the processing time and storage for retrieving and indexing.
Design and Realization of Music Retrieval System Based on Feature Content
