Recognition of convolutive speech mixtures by missing feature techniques for ICA

Dorothea Kolossa*, Hiroshi Sawada, Ramon Fernandez Astudillo, Reinhold Orglmeister, Shoji Makino

*この研究の対応する著者

研究成果

8 被引用数 (Scopus)

抄録

One challenging problem for robust speech recognition is the cocktail party effect, where multiple speaker signals are active simultaneously in an overlapping frequency range. In that case, independent component analysis (ICA) can separate the signals in reverberant environments, also. However, incurred feature distortions prove detrimental for speech recognition. To reduce consequential recognition errors, we describe the use of ICA for the additional estimation of uncertainty information. This information is subsequently used in missing feature speech recognition, which leads to far more correct and accurate recognition also in reverberant situations at RT60 = 300ms.

本文言語English
ホスト出版物のタイトルConference Record of the 40th Asilomar Conference on Signals, Systems and Computers, ACSSC '06
ページ1397-1401
ページ数5
DOI
出版ステータスPublished - 2006
外部発表はい
イベント40th Asilomar Conference on Signals, Systems, and Computers, ACSSC '06 - Pacific Grove, CA, United States
継続期間: 2006 10 292006 11 1

出版物シリーズ

名前Conference Record - Asilomar Conference on Signals, Systems and Computers
ISSN(印刷版)1058-6393

Conference

Conference40th Asilomar Conference on Signals, Systems, and Computers, ACSSC '06
国/地域United States
CityPacific Grove, CA
Period06/10/2906/11/1

ASJC Scopus subject areas

  • 信号処理
  • コンピュータ ネットワークおよび通信

フィンガープリント

「Recognition of convolutive speech mixtures by missing feature techniques for ICA」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