no code implementations • 14 May 2024 • Andreas Brendel, Nicola Pia, Kishan Gupta, Guillaume Fuchs, Markus Multrus
This discrete representation is typically generated by applying a quantizer to the output of the neural encoder.
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 • 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).
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).
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 • 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.