Experimental Study of Generalized Subspace Filters for the Cocktail Party Situation

Knud Bank Christensen, Mads Græsbøll Christensen, Jesper B. Boldt, Fredrik Gran

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

4 Citations (Scopus)
71 Downloads (Pure)

Abstract

This paper investigates the potential performance of generalized
subspace filters for speech enhancement in cocktail party
situations with very poor signal/noise ratio, e.g. down to
-15 dB. Performance metrics output signal/noise ratio, signal/
distortion ratio, speech quality rating and speech intelligibility
rating are mapped as functions of two algorithm parameters,
revealing clear trade-off options between noise, distortion
and subjective performances and a recommended choice
of trade-off. Given sufficiently good noise statistics, SNR
improvements around 20 dB as well as PESQ quality and
STOI intelligibility rating improvements exceeding 1.0 and
0.2 points respectively are found. This shows the potential of
the method.
Original languageEnglish
Title of host publicationProceedings IEEE ICASSP 2016
Number of pages5
PublisherIEEE
Publication date2016
DOIs
Publication statusPublished - 2016
EventThe 41st IEEE International Conference on Acoustics, Speech and Signal Processing - Shanghai, China
Duration: 20 Mar 201625 Mar 2016
http://www.icassp2016.org/

Conference

ConferenceThe 41st IEEE International Conference on Acoustics, Speech and Signal Processing
Country/TerritoryChina
CityShanghai
Period20/03/201625/03/2016
Internet address
SeriesI E E E International Conference on Acoustics, Speech and Signal Processing. Proceedings
ISSN1520-6149

Keywords

  • Speech enhancement, subspace signal processing, optimal filtering, babble noise, cocktail party

Fingerprint

Dive into the research topics of 'Experimental Study of Generalized Subspace Filters for the Cocktail Party Situation'. Together they form a unique fingerprint.

Cite this