A two-stage frequency-domain blind source separation method for underdetermined convolutive mixtures

Hiroshi Sawada*, Shoko Araki, Shoji Makino

*この研究の対応する著者

研究成果: Conference contribution

68 被引用数 (Scopus)

抄録

This paper proposes a two-stage method for the blind separation of convolutively mixed sources. We employ time-frequency masking, which can be applied even to an underdetermined case where the number of sensors is insufficient for the number of sources. In the first stage of the method, frequency bin-wise mixtures are classified based on Gaussian mixture model fitting. In the second stage, the permutation ambiguities of the bin-wise classified signals are aligned by clustering the posterior probability sequences calculated in the first stage. Experimental results for separating four speeches with three microphones under reverberant conditions show the superiority of the proposed method over existing methods based on time-difference-of-arrival estimations or signal envelope clustering.

本文言語English
ホスト出版物のタイトル2007 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA
ページ139-142
ページ数4
DOI
出版ステータスPublished - 2007
外部発表はい
イベント2007 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA - New Paltz, NY, United States
継続期間: 2007 10 212007 10 24

出版物シリーズ

名前IEEE Workshop on Applications of Signal Processing to Audio and Acoustics

Conference

Conference2007 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, WASPAA
国/地域United States
CityNew Paltz, NY
Period07/10/2107/10/24

ASJC Scopus subject areas

  • 信号処理

フィンガープリント

「A two-stage frequency-domain blind source separation method for underdetermined convolutive mixtures」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