Optimal Time-Domain Noise Reduction Filters: A Theoretical Study

Jacob Benesty, Jingdong Chen

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

9 Scopus citations

Abstract

Additive noise is ubiquitous in acoustics environments and can affect the intelligibility and quality of speech signals. Therefore, a so-called noise reduction algorithm is required to mitigate the effect of the noise that is picked up by the microphones. In this work, we propose a general framework in the time domain for the single and multiple microphone cases, from which it is very convenient to derive, study, and analyze all kind of optimal noise reduction filters. Not only all known algorithms can be deduced from this approach, shedding more light on how they function, but new ones can be discovered as well.

Original languageEnglish
Title of host publicationSpringerBriefs in Speech Technology
PublisherSpringer Science and Business Media B.V.
Pages1-75
Number of pages75
DOIs
StatePublished - 2011

Publication series

NameSpringerBriefs in Speech Technology
Volume1
ISSN (Print)2191-737X
ISSN (Electronic)2191-7388

Keywords

  • LCMV filter
  • maximun SNR filter
  • MDVR filter
  • microphone arrays
  • multichannel
  • noise reduction
  • prediction filter
  • single-channel
  • space-time prediction filter
  • speech enhancement
  • time domain
  • tradeoff filter
  • Wiener filter

Fingerprint

Dive into the research topics of 'Optimal Time-Domain Noise Reduction Filters: A Theoretical Study'. Together they form a unique fingerprint.

Cite this