Maximum negentropy beamforming with superdirectivity

K. Kumatani, L. Lu, J. McDonough, A. Ghoshal, D. Klakow

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents new superdirective beamforming algorithms based on the maximum negentropy (MN) criterion for distant automatic speech recognition. The MN beamformer is configured in the generalized sidelobe canceler structure, and uses the weights derived from a delay-and-sum beamformer as the quiescent weight vector. While satisfying the distortionless constraint in the look direction, it adjusts the active weight vector to make the output maximally super-Gaussian. The current paper proposes to use the weights of a superdirective beamformer as the quiescent vector, which results in improved directivity and noise suppression at lower frequencies. We demonstrate the effectiveness of our approach through far-field speech recognition experiments on the Multi-Channel Wall Street Journal Audio Visual Corpus (MC-WSJ-AV). The technique proposed in the current paper reduces the word error rate (WER) by 56% relative to a single distant microphone baseline, which is a 14% reduction in WER over the original MN beamformer formulation.
Original languageEnglish
Title of host publicationProceedings of the 18th European Signal Processing Conference EUSIPCO 2010
EditorsBastiaan Kleijn, Jan Larsen
Place of PublicationKessariani, Greece
PublisherEuropean Association for Signal, Speech, and Image Processing (EURASIP)
Pages2067-2071
Number of pages5
Publication statusPublished - 1 Aug 2010
Event18th European Signal Processing Conference (EUSIPCO 2010) - Aalborg, Denmark
Duration: 23 Aug 201027 Aug 2010

Publication series

NameSignal Processing Conference, 2010 18th European
PublisherIEEE
ISSN (Print)2219-5491
NameProceedings of the 18th European Signal Processing Conference EUSIPCO-2010
PublisherEuropean Association for Signal, Speech, and Image Processing (EURASIP),
ISSN (Print)2076-1465

Conference

Conference18th European Signal Processing Conference (EUSIPCO 2010)
Country/TerritoryDenmark
CityAalborg
Period23/08/1027/08/10

Keywords

  • array signal processing
  • entropy
  • microphones
  • speech recognition
  • MC-WSJ-AV
  • WER
  • delay-and-sum beamformer
  • distant automatic speech recognition
  • maximum negentropy beamforming
  • multichannel wall street journal audio visual corpus
  • noise suppression
  • quiescent vector
  • quiescent weight vector
  • sidelobe canceler structure
  • single distant microphone baseline
  • superdirective beamformer
  • superdirective beamforming
  • word error rate
  • Array signal processing
  • Arrays
  • Entropy
  • Microphones
  • Noise
  • Speech
  • Speech recognition

Fingerprint

Dive into the research topics of 'Maximum negentropy beamforming with superdirectivity'. Together they form a unique fingerprint.

Cite this