Skip to main navigation Skip to search Skip to main content

Speaker extraction using LCMV beamformer with DNN-based SPP and RTF identification scheme

  • Ariel Malek
  • , Shlomo E. Chazan
  • , Ilan Malka
  • , Vladimir Tourbabin
  • , Jacob Goldberger
  • , Eli Tzirkel-Hancock
  • , Sharon Gannot

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

The linearly constrained minimum variance (LCMV)-beamformer (BF) is a viable solution for desired source extraction from a mixture of speakers in a noisy environment. The performance in terms of speech distortion, interference cancellation and noise reduction depends on the estimation of a set of parameters. This paper presents a new mechanism to update the parameters of the LCMV-BF. A new speech presence probability (SPP)-based voice activity detector (VAD) controls the noise covariance matrix update, and a speaker position identifier (SPI) procedure controls the relative transfer functions (RTFs) update. A postfilter is then applied to the BF output to further attenuate the residual noise signal. A series of experiments using real-life recordings confirm the speech enhancement capabilities of the proposed algorithm.

Original languageEnglish
Title of host publication25th European Signal Processing Conference, EUSIPCO 2017
PublisherInstitute of Electrical and Electronics Engineers
Pages2274-2278
Number of pages5
ISBN (Electronic)9780992862671
DOIs
StatePublished - 23 Oct 2017
Externally publishedYes
Event25th European Signal Processing Conference, EUSIPCO 2017 - Kos, Greece
Duration: 28 Aug 20172 Sep 2017

Publication series

Name25th European Signal Processing Conference, EUSIPCO 2017
Volume2017-January

Conference

Conference25th European Signal Processing Conference, EUSIPCO 2017
Country/TerritoryGreece
CityKos
Period28/08/172/09/17

ASJC Scopus subject areas

  • Signal Processing

Fingerprint

Dive into the research topics of 'Speaker extraction using LCMV beamformer with DNN-based SPP and RTF identification scheme'. Together they form a unique fingerprint.

Cite this