An expectation-maximization algorithm for multichannel adaptive speech dereverberation in the frequency-domain

• Published in: 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 17 - 20
• Main Authors: Schmid, D., Malik, S., Enzner, G.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2012
• Subjects: Expectation-maximization; Frequency modulation; frequency-domain adaptive filtering; Microphones; multichannel dereverberation; Noise; Reverberation; Speech; state-space model; Vectors

This paper presents an online dereverberation algorithm that is derived within the maximum-likelihood expectation-maximization (ML-EM) framework. We formulate an overlap-save observation model for the multichannel blind problem in the DFT-domain. The modeling of acoustic channel impulse responses as...