Jointly optimal near-end and far-end multi-microphone speech intelligibility enhancement based on mutual information

Conference paper (2016)

Authors

S. Khademi Signal Processing Systems - EEMCS
R.C. Hendriks Signal Processing Systems - EEMCS
W.B. Kleijn Signal Processing Systems - EEMCS
Research Group
Signal Processing Systems (EEMCS) (TU Delft)
Copyright: © 2016 S. Khademi, R.C. Hendriks, W.B. Kleijn
DOI
help_outline
https://doi.org/10.1109/icassp.2016.7471756
Speech enhancement Speech intelligibility enhancement Mutal information Signal to noise ratio Roduction Multi-microphone Minimum variance distortionless response (MVDR) Beamformer
More Info
expand_more

Published Date

19-05-2016

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Department

Microelectronics

Research Group

Signal Processing Systems

Abstract

The processing required for the global maximization of the intelligibility of speech acquired by multiple microphones and rendered by a single loudspeaker, is considered in this paper. The intelligibility is quantized, based on the mutual information rate between the message spoken by the talker and the message as interpreted by the listener. We prove that then, in each of a set of narrow-band channels, the processing can be decomposed into a minimum variance distortionless response (MVDR) beamforming operation that reduces the noise in the talker environment, followed by a gain operation that, given the far-end noise and beamforming operation, accounts for the noise at the listener end. Our experiments confirm that both processing steps are necessary for the effective conveyance ofa message and, importantly, that the second step must be aware of the first step.