Audio source segmentation using spectral correlation features for automatic indexing of broadcast news

Shoichi Matsunaga, Osamu Mizuno, Katsutoshi Ohtsuki, Yoshihiko Hayashi

研究成果: Conference contribution

1 被引用数 (Scopus)

抄録

This paper proposes a new segmentation procedure to detect audio source intervals for automatic indexing of broadcast news. The procedure is composed of an audio source detection part and a part that smoothes the detected sequences. The detection part uses three new acoustic feature parameters that are based on spectral cross-correlation: spectral stability, white noise similarity, and sound spectral shape. These parameters make it possible to capture the audio sources more accurately than can be done with conventional parameters. The smoothing part has a new merging method that drops erroneous detection results of short duration. Audio source classification experiments are conducted on broadcast news segments. Performance is increased by 6.6% when the proposed parameters are used and by 3.1% when the proposed merging method is used, showing the usefulness of our approach. Experiments confirm the impact of this proposal on broadcast news indexing.

本文言語English
ホスト出版物のタイトル2004 12th European Signal Processing Conference, EUSIPCO 2004
出版社European Signal Processing Conference, EUSIPCO
ページ2103-2106
ページ数4
ISBN(電子版)9783200001657
出版ステータスPublished - 2015 4月 3
外部発表はい
イベント12th European Signal Processing Conference, EUSIPCO 2004 - Vienna, Austria
継続期間: 2004 9月 62004 9月 10

出版物シリーズ

名前European Signal Processing Conference
06-10-September-2004
ISSN(印刷版)2219-5491

Conference

Conference12th European Signal Processing Conference, EUSIPCO 2004
国/地域Austria
CityVienna
Period04/9/604/9/10

ASJC Scopus subject areas

  • 信号処理
  • 電子工学および電気工学

フィンガープリント

「Audio source segmentation using spectral correlation features for automatic indexing of broadcast news」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル