Spectral feature-based nonlinear residual echo suppression

Schwarz A, Hofmann C, Kellermann W (2013)


Publication Language: English

Publication Status: Published

Publication Type: Conference contribution, Conference Contribution

Publication year: 2013

Pages Range: 1-4

Article Number: 6701825

Event location: New Paltz, NY US

ISBN: 978-1-4799-0972-8

DOI: 10.1109/WASPAA.2013.6701825

Abstract

We propose a method for nonlinear residual echo suppression that consists of extracting spectral features from the far-end signal, and using an artificial neural network to model the residual echo magnitude spectrum from these features. We compare the modeling accuracy achieved by realizations with different features and network topologies, evaluating the mean squared error of the estimated residual echo magnitude spectrum. We also present a low complexity real-time implementation combining an offline-trained network with online adaptation, and investigate its performance in terms of echo suppression and speech distortion for real mobile phone recordings. © 2013 IEEE.

Authors with CRIS profile

How to cite

APA:

Schwarz, A., Hofmann, C., & Kellermann, W. (2013). Spectral feature-based nonlinear residual echo suppression. In Proceedings of the IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA) (pp. 1-4). New Paltz, NY, US.

MLA:

Schwarz, Andreas, Christian Hofmann, and Walter Kellermann. "Spectral feature-based nonlinear residual echo suppression." Proceedings of the IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), New Paltz, NY 2013. 1-4.

BibTeX: Download