Site menu:

Dual-microphone speech dereverberation
in a noisy environment

E.A.P. Habets, S. Gannot and I. Cohen

Abstract

Speech signals recorded with a distant microphone usually contain reverberation and noise, which degrade the fidelity and intelligibility of speech, and the recognition performance of automatic speech recognition systems. In [1] we presented a multi-channel speech dereverberation algorithm to suppress late reverberation in a noise free environment. In this paper we show how an estimate of the late reverberant energy can be obtained from noisy observations. A more sophisticated speech enhancement technique based on the Optimally-Modified Log Spectral Amplitude (OM-LSA) estimator is used to suppress the undesired late reverberant signal and noise. The speech presence probability used in the OM-LSA is extended to improve the decision between speech, late reverberation and noise. Experiments using simulated and real acoustic impulse responses are presented and show significant reverberation reduction with little speech distortion.

References

  1. Habets, E.A.P.
    Multi-Channel Speech Dereverberation based on a Statistical Model of Late Reverberation, Proc. ICASSP 2005, 30th IEEE International Conference on Acoustics, Speech, and Signal Processing, Philadelphia, USA, March 18-23, 2005, pp. 173-176.

Results

Room A @ 1m (Using measured Room Impulse Responses)
Microphone
Microphone (noise-free)
Delay & Sum Beamformer
Proposed with Direct Path Compensation
Proposed without Direct Path Compensation

Room A @ 2m (Using measured Room Impulse Responses)
Microphone
Microphone (noise-free)
Delay & Sum Beamformer
Proposed with Direct Path Compensation
Proposed without Direct Path Compensation

Room B @ 1m (Using simulated Room Impulse Responses)
Microphone
Microphone (noise-free)
Delay & Sum Beamformer
Proposed with Direct Path Compensation
Proposed without Direct Path Compensation

Room B @ 2m (Using simulated Room Impulse Responses)
Microphone
Microphone (noise-free)
Delay & Sum Beamformer
Proposed with Direct Path Compensation
Proposed without Direct Path Compensation

Note: It is strongly recommended to use headphones for listening tests.

Status

Published in the Proc. of the IEEE International Symposium on Signal Processing and Information Technology (ISSPIT 2006), Vancouver, Canada, Aug. 27-30, 2006, pp. 651-655.

BibTex Entry

@INPROCEEDINGS{Habets2006b,
  author = {E. A. P. Habets and S. Gannot and I. Cohen},
  title = {Dual-Microphone Speech Dereverberation in a Noisy Environment},
  booktitle = {Proc. {IEEE} Intl. Symposium on Signal Processing and Information
	Technology (ISSPIT)},
  year = {2006},
  pages = {651--655},
  address = {Vancouver, Canada},
  month = aug
}

[Back]