A convolutive spectral decomposition approach to the separation of feedback from target speech

Gautham J. Mysore, Paris Smaragdis

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

Abstract

Feedback is a common problem in teleconferencing systems. Typical usage of an adaptive filter can be effective for feedback reduction but it relies on the presence of such a filter on the side of the far speaker in order to reduce feedback on the side of the near speaker. In order to avoid this reliance on the far speaker's setup, we can use an adaptive filter on the side of the near speaker. Unfortunately, due to non-linear speech coding typically used during speech transmission, these filters perform poorly in this situation. In this paper, we present a novel probabilistic method, using a non-negative convolutive decomposition of spectrogram data to perform feedback reduction by posing the problem as a source separation problem. Our method is robust to non-linear speech coding as well as continuous double-talk, which often presents a challenge to adaptive filters. We compare our method to the use of an adaptive filter and show superior results with respect to standard source separation metrics.

Original languageEnglish (US)
Title of host publication2011 IEEE International Workshop on Machine Learning for Signal Processing - Proceedings of MLSP 2011
DOIs
StatePublished - 2011
Event21st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2011 - Beijing, China
Duration: Sep 18 2011Sep 21 2011

Publication series

NameIEEE International Workshop on Machine Learning for Signal Processing

Other

Other21st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2011
Country/TerritoryChina
CityBeijing
Period9/18/119/21/11

Keywords

  • Feedback Reduction
  • Non-Negative Spectrogram Factorization
  • Source Separation

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Signal Processing

Fingerprint

Dive into the research topics of 'A convolutive spectral decomposition approach to the separation of feedback from target speech'. Together they form a unique fingerprint.

Cite this