1 code implementation • 4 Jun 2023 • Thomas Haubner, Andreas Brendel, Walter Kellermann
To obtain precise echo estimates, the parameters of the echo canceler, i. e., the filter coefficients, need to be estimated quickly and precisely from the observed loudspeaker and microphone signals.
no code implementations • 14 Apr 2023 • Matthias Kreuzer, Walter Kellermann
In this article, we present our contribution to the ICPHM 2023 Data Challenge on Industrial Systems' Health Monitoring using Vibration Analysis.
no code implementations • 14 Apr 2023 • Matthias Kreuzer, David Schmidt, Simon Wokusch, Walter Kellermann
In this paper, we address the challenging problem of detecting bearing faults in railway vehicles by analyzing acoustic signals recorded during regular operation.
no code implementations • 14 Apr 2023 • Matthias Kreuzer, Alexander Schmidt, Walter Kellermann
However, these features are usually evaluated on data originating from relatively simple scenarios and a significant performance loss can be observed if more realistic scenarios are considered.
no code implementations • 14 Mar 2023 • Annika Briegleb, Thomas Haubner, Vasileios Belagiannis, Walter Kellermann
Beamforming for multichannel speech enhancement relies on the estimation of spatial characteristics of the acoustic scene.
no code implementations • 27 Oct 2022 • Annika Briegleb, Mhd Modar Halimeh, Walter Kellermann
In conventional multichannel audio signal enhancement, spatial and spectral filtering are often performed sequentially.
no code implementations • 28 Jul 2022 • Andreas Brendel, Thomas Haubner, Walter Kellermann
Most of the currently used algorithms belong to one of the following three families: Frequency Domain ICA (FD-ICA), Independent Vector Analysis (IVA), and TRIple-N Independent component analysis for CONvolutive mixtures (TRINICON).
1 code implementation • 13 May 2022 • Thomas Haubner, Zbyněk Koldovský, Walter Kellermann
We describe a joint acoustic echo cancellation (AEC) and blind source extraction (BSE) approach for multi-microphone acoustic frontends.
no code implementations • 3 Mar 2022 • Thomas Haubner, Walter Kellermann
We introduce a novel method for controlling the functionality of a hands-free speech communication device which comprises a model-based acoustic echo canceller (AEC), minimum variance distortionless response (MVDR) beamformer (BF) and spectral postfilter (PF).
no code implementations • 24 Jan 2022 • Michael Günther, Andreas Brendel, Walter Kellermann
In this contribution, we provide a comprehensive analysis of model-based microphone utility estimation approaches that use signal features and, as an alternative, also propose machine learning-based estimation methods that identify optimal sensor signal utility features.
no code implementations • 5 Oct 2021 • Andreas Brendel, Johannes Zeitler, Walter Kellermann
Many spatial filtering algorithms used for voice capture in, e. g., teleconferencing applications, can benefit from or even rely on knowledge of Relative Transfer Functions (RTFs).
1 code implementation • 6 Aug 2021 • Mhd Modar Halimeh, Walter Kellermann
As shown by the experimental results, the proposed approach is able to exploit both spatial and spectral characteristics of the desired source signal resulting in a physically plausible spatial selectivity and superior speech quality compared to other baseline methods.
no code implementations • 2 Jun 2021 • Thomas Haubner, Andreas Brendel, Walter Kellermann
We present a novel end-to-end deep learning-based adaptation control algorithm for frequency-domain adaptive system identification.
no code implementations • 7 May 2021 • Thomas Haubner, Andreas Brendel, Walter Kellermann
The proposed method assumes that the variability of AIRs of an acoustic scene is confined to a low-dimensional manifold which is embedded in a high-dimensional space of possible AIR estimates.
no code implementations • 16 Dec 2020 • Thomas Haubner, Mhd. Modar Halimeh, Andreas Brendel, Walter Kellermann
We introduce a synergistic approach to double-talk robust acoustic echo cancellation combining adaptive Kalman filtering with a deep neural network-based postfilter.
no code implementations • 31 Jul 2020 • Cornelius Frankenbach, Pablo Martínez-Nuevo, Martin Møller, Walter Kellermann
In particular, we propose an iterative method to reconstruct bandlimited multidimensional signals based on truncated versions of the original signal to bounded regions---herein referred to as continuous measurements.
no code implementations • 20 Mar 2020 • Andreas Brendel, Walter Kellermann
Algorithms for Blind Source Separation (BSS) of acoustic signals require efficient and fast converging optimization strategies to adapt to nonstationary signal statistics and time-varying acoustic scenarios.
1 code implementation • 3 Sep 2019 • Christine Evers, Heinrich Loellmann, Heinrich Mellmann, Alexander Schmidt, Hendrik Barfuss, Patrick Naylor, Walter Kellermann
The aim of the LOCAlization and TrAcking (LOCATA) Challenge is an open-access framework for the objective evaluation and benchmarking of broad classes of algorithms for sound source localization and tracking.
no code implementations • 4 Aug 2016 • Christian Huemmer, Ramón Fernández Astudillo, Walter Kellermann
In this paper, we advance a recently-proposed uncertainty decoding scheme for DNN-HMM (deep neural network - hidden Markov model) hybrid systems.
no code implementations • 14 Apr 2016 • Christian Huemmer, Christian Hofmann, Roland Maas, Walter Kellermann
In this article, we present the elitist particle filter based on evolutionary strategies (EPFES) as an efficient approach for nonlinear system identification.
1 code implementation • 12 Feb 2015 • Andreas Schwarz, Walter Kellermann
Several novel unbiased CDR estimators are proposed, and it is shown that knowledge of either the direction of arrival (DOA) of the target source or the coherence of the noise field is sufficient for unbiased CDR estimation.
Sound
no code implementations • 18 Nov 2014 • Christian Huemmer, Roland Maas, Walter Kellermann
In this article, we derive a new stepsize adaptation for the normalized least mean square algorithm (NLMS) by describing the task of linear acoustic echo cancellation from a Bayesian network perspective.
no code implementations • 9 Oct 2014 • Andreas Schwarz, Christian Huemmer, Roland Maas, Walter Kellermann
We propose a spatial diffuseness feature for deep neural network (DNN)-based automatic speech recognition to improve recognition accuracy in reverberant and noisy environments.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 11 Oct 2013 • Roland Maas, Christian Huemmer, Armin Sehr, Walter Kellermann
This article provides a unifying Bayesian network view on various approaches for acoustic model adaptation, missing feature, and uncertainty decoding that are well-known in the literature of robust automatic speech recognition.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2