Site menu:

Signal-based Performance Evaluation of Dereverberation Algorithms

P.A. Naylor, N.D. Gaubitch and E.A.P. Habets

Abstract

We address the measurement of reverberation in terms of the Direct to Reverberant Ratio (DRR) in the context of the assessment of dereverberation algorithms for which we wish to quantify the level of reverberation before and after processing. The DRR is normally calculated from the impulse response of the reverberating system. However, several important dereverberation algorithms involve nonlinear and/or time-varying processing and therefore their effect cannot conveniently be represented in terms of modifications to the impulse response of the reverberating system. In such cases, we show that a good estimate of DRR can be obtained from the input/output signals alone using the Signal to Reverberant Ratio (SRR) only if the source signal is spectrally white and correctly normalized. We study alternative normalization schemes and conclude by showing a least squares optimal normalization procedure for estimating DRR using signal-based SRR measurement. Simulation results illustrate the accuracy of DRR estimation using SRR.

Status

Published in the Journal of Electrical and Computer Engineering (formerly Hindawi Research Letters in Signal Processing), 2010.

Download

A MATLAB implementation of the proposed signal-based quality measure is available here. Note that the proposed function requires Voicebox, a speech processing toolbox for MATLAB maintained by and mostly written by Mike Brookes from Imperial College London.

BibTex Entry

@ARTICLE{Naylor2010,
  author = {P.A. Naylor and N.D. Gaubitch and E.A.P. Habets},
  title = {Signal-based Performance Evaluation of Dereverberation Algorithms},
  journal = {Journal of Electrical and Computer Engineering},
  year = {2010},
  notes = {Article ID 127513},
  volume = 2010,
  pages = 1--5,
}

[Back]