Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Non-intrusive regularization for least-squares multichannel equalization for speech dereverberation

: Kodrasi, I.; Doclo, S.; Goetze, S.


Institute of Electrical and Electronics Engineers -IEEE-:
27th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2012. Vol.2 : 14 - 17 Nov. 2012, Eilat / IEEE Israel Section
Piscataway: IEEE, 2012
ISBN: 978-1-4673-4682-5
ISBN: 978-1-4673-4680-1 (print)
ISBN: 978-1-4673-4681-8
IEEE Convention of Electrical & Electronics Engineers in Israel (IEEEI) <27, 2012, Eilat>
Conference Paper
Fraunhofer IDMT ()

Acoustic multichannel equalization techniques for speech dereverberation are known to be highly sensitive to estimation errors of the room impulse responses. In order to increase robustness, it has been proposed to incorporate regularization. However, the optimal regularization parameter which yields the highest perceptual speech quality has generally been determined intrusively, limiting the practical applicability. In this paper, we propose an automatic non-intrusive procedure for determining the regularization parameter based on the L-curve. Experimental results show that using such an automatic non-intrusive regularization parameter in a recently proposed partial multichannel equalization technique (P-MINT) leads to a very similar performance as using the intrusively determined optimal regularization parameter. Furthermore, it is shown that the automatically regularized P-MINT technique outperforms state-of-the-art multichannel equalization techniques such as channe l shortening and relaxed multichannel least-squares, both in terms of reverberant tail suppression and perceptual speech quality.