• DocumentCode
    2936388
  • Title

    Query by rhythm: an approach for song retrieval in music databases

  • Author

    Chen, James C C ; Chen, Arbee L P

  • Author_Institution
    Dept. of Comput. Sci., Nat. Tsing Hua Univ., Hsinchu, Taiwan
  • fYear
    1998
  • fDate
    23-24 Feb 1998
  • Firstpage
    139
  • Lastpage
    146
  • Abstract
    We propose techniques for retrieving songs by rhythm from music databases. The rhythm of songs is modeled by rhythm strings. The song retrieval problem is then transformed to the string matching problem. In order to allow approximate string matching, we define similarity measures on rhythm strings. An index structure, called L-tree, is proposed to support efficient sub-string matching. Retrieval algorithms based on L-tree are then designed to provide approximate and sub- song retrieval. Experimental results show that this approach is effective and efficient
  • Keywords
    database management systems; multimedia computing; music; query processing; string matching; L-tree; index structure; music databases; query by rhythm; retrieval algorithms; rhythm strings; similarity measures; song retrieval; string matching; substring matching; Audio databases; Computer networks; Computer science; Data structures; Image databases; Multiple signal classification; Music information retrieval; Read only memory; Rhythm; Spatial databases;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Research Issues In Data Engineering, 1998. 'Continuous-Media Databases and Applications'. Proceedings., Eighth International Workshop on
  • Conference_Location
    Orlando, FL
  • Print_ISBN
    0-8186-8389-9
  • Type

    conf

  • DOI
    10.1109/RIDE.1998.658288
  • Filename
    658288