事例に基づく演奏表情の生成  [in Japanese] A Case-based Approach to the Generation of Musical Expression  [in Japanese]

Access this Article

Search this Article

Author(s)

    • 鈴木 泰山 SUZUKI Taizan
    • 東京工業大学情報理工学研究科計算工学専攻 Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology
    • 徳永 健伸 TOKUNAGA Takenobu
    • 東京工業大学情報理工学研究科計算工学専攻 Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology
    • 田中 穂積 TANAKA Hozumi
    • 東京工業大学情報理工学研究科計算工学専攻 Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology

Abstract

本論文では,事例に基づいて演奏表情を生成する手法を提案し,この手法を用いたシステムによる実験について報告する.人間による音楽演奏は,テンポや音の強さの揺らぎなどの演奏表情と呼ばれる要素を含んでいる.演奏表情の与え方は自由度が非常に高く,演奏者の経験や演奏スタイル,演奏時の気分などの演奏状況に応じて多様に変化する.しかしながら,従来の演奏表情の自動生成に関する研究では演奏状況をほとんど考慮していないため,演奏状況に応じた演奏表情の生成をすることができなかった.我々は演奏状況に応じた多様な演奏表情を生成するために,事例に基づく演奏表情の生成を提案する.この手法は過去の演奏データの中から,演奏する楽曲に類似した楽曲を探し,その楽曲に付与された演奏表情を利用して対象曲に演奏表情を付与する.我々はこの手法を利用した演奏表情生成システムを実装し,演奏状況に応じた演奏表情の生成を試みた.その結果,この手法が多様な演奏表情の生成に有効であることを確認した.This paper presents a case-based formulation of computationalmusical expression generation, and reports the results ofexperiments on the derived musical expression generation system.The majority of musical performance has musical expression,including such items as fluctuation in tempo and volume.Generally, musical expression is freely attributed by the performer,and is affected by various conditions surrounding performance(performance conditions), such as the performer's taste, performativestyle, mood, and so forth.However, past research on the automatic generation of musicalexpression has downplayed performance conditions.In this paper, we propose a case-based approach to the automaticgeneration of musical expression.This method easily enables the generation of musical expressionreflecting the performance conditions.We have implemented the proposed method in a performance datageneration system, and generated some expressive performance data fromthis system.According to our evaluation of generated performance data, our methodwas shown to be effective at generating musical expression customizedto performance conditions.

This paper presents a case-based formulation of computational musical expression generation, and reports the results of experiments on the derived musical expression generation system. The majority of musical performance has musical expression, including such items as fluctuation in tempo and volume. Generally, musical expression is freely attributed by the performer, and is affected by various conditions surrounding performance (performance conditions), such as the performer's taste, performative style, mood, and so forth. However, past research on the automatic generation of musical expression has downplayed performance conditions. In this paper, we propose a case-based approach to the automatic generation of musical expression. This method easily enables the generation of musical expression reflecting the performance conditions. We have implemented the proposed method in a performance data generation system, and generated some expressive performance data from this system. According to our evaluation of generated performance data, our method was shown to be effective at generating musical expression customized to performance conditions.

Journal

  • Transactions of Information Processing Society of Japan

    Transactions of Information Processing Society of Japan 41(4), 1134-1145, 2000-04-15

    Information Processing Society of Japan (IPSJ)

References:  12

Cited by:  15

Codes

  • NII Article ID (NAID)
    110002725321
  • NII NACSIS-CAT ID (NCID)
    AN00116647
  • Text Lang
    JPN
  • Article Type
    Journal Article
  • ISSN
    1882-7764
  • NDL Article ID
    5345018
  • NDL Source Classification
    ZM13(科学技術--科学技術一般--データ処理・計算機)
  • NDL Call No.
    Z14-741
  • Data Source
    CJP  CJPref  NDL  NII-ELS  IPSJ 
Page Top