Speaker Change Detection Based on a Weighted Distance Measure over the Centroid Model

この論文にアクセスする

この論文をさがす

著者

    • SEO Jin Soo
    • the Department of Electrical Engineering, Gangneung-Wonju National University

抄録

Speaker change detection involves the identification of the time indices of an audio stream, where the identity of the speaker changes. This paper proposes novel measures for speaker change detection over the centroid model, which divides the feature space into non-overlapping clusters for effective speaker-change comparison. The centroid model is a computationally-efficient variant of the widely-used mixture-distribution based background models for speaker recognition. Experiments on both synthetic and real-world data were performed; the results show that the proposed approach yields promising results compared with the conventional statistical measures.

収録刊行物

  • IEICE transactions on information and systems

    IEICE transactions on information and systems 95(5), 1543-1546, 2012-05-01

    一般社団法人 電子情報通信学会

参考文献:  9件中 1-9件 を表示

各種コード

  • NII論文ID(NAID)
    10030943373
  • NII書誌ID(NCID)
    AA10826272
  • 本文言語コード
    ENG
  • 資料種別
    SHO
  • ISSN
    09168532
  • データ提供元
    CJP書誌  J-STAGE 
ページトップへ