Source author record

Sebastian J. Schlecht

Sebastian J. Schlecht appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

eess.AS Sound eess.SP eess.SY Machine Learning physics.comp-ph Systems and Control

Catalog footprint

What is connected

4works

7topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

AnyRIR: Robust Non-intrusive Room Impulse Response Estimation in the Wild

We address the problem of estimating room impulse responses (RIRs) in noisy, uncontrolled environments where non-stationary sounds such as speech or footsteps corrupt conventional deconvolution. We propose AnyRIR, a non-intrusive method that uses music as the excitation signal instead of a dedicated test signal, and formulate RIR estimation as an L1-norm regression in the time-frequency domain. Solved efficiently with Iterative Reweighted Least Squares (IRLS) and Least-Squares Minimal Residual (LSMR) methods, this approach exploits the sparsity of non-stationary noise to suppress its influence. Experiments on simulated and measured data show that AnyRIR outperforms L2-based and frequency-domain deconvolution, under in-the-wild noisy scenarios and codec mismatch, enabling robust RIR estimation for AR/VR and related applications.

preprint2022arXiv

Physical Modeling using Recurrent Neural Networks with Fast Convolutional Layers

Discrete-time modeling of acoustic, mechanical and electrical systems is a prominent topic in the musical signal processing literature. Such models are mostly derived by discretizing a mathematical model, given in terms of ordinary or partial differential equations, using established techniques. Recent work has applied the techniques of machine-learning to construct such models automatically from data for the case of systems which have lumped states described by scalar values, such as electrical circuits. In this work, we examine how similar techniques are able to construct models of systems which have spatially distributed rather than lumped states. We describe several novel recurrent neural network structures, and show how they can be thought of as an extension of modal techniques. As a proof of concept, we generate synthetic data for three physical systems and show that the proposed network structures can be trained with this data to reproduce the behavior of these systems.

preprint2021arXiv

Allpass Feedback Delay Networks

In the 1960s, Schroeder and Logan introduced delay line-based allpass filters, which are still popular due to their computational efficiency and versatile applicability in artificial reverberation, decorrelation, and dispersive system design. In this work, we extend the theory of allpass systems to any arbitrary connection of delay lines, namely feedback delay networks (FDNs). We present a characterization of uniallpass FDNs, i.e., FDNs, which are allpass for an arbitrary choice of delays. Further, we develop a solution to the completion problem, i.e., given an FDN feedback matrix to determine the remaining gain parameters such that the FDN is allpass. Particularly useful for the completion problem are feedback matrices, which yield a homogeneous decay of all system modes. Finally, we apply the uniallpass characterization to previous FDN designs, namely, Schroeder's series allpass and Gardner's nested allpass for single-input, single-output systems, and, Poletti's unitary reverberator for multi-input, multi-output systems and demonstrate the significant extension of the design space.

preprint2020arXiv

Scattering in Feedback Delay Networks

Feedback delay networks (FDNs) are recursive filters, which are widely used for artificial reverberation and decorrelation. One central challenge in the design of FDNs is the generation of sufficient echo density in the impulse response without compromising the computational efficiency. In a previous contribution, we have demonstrated that the echo density of an FDN can be increased by introducing so-called delay feedback matrices where each matrix entry is a scalar gain and a delay. In this contribution, we generalize the feedback matrix to arbitrary lossless filter feedback matrices (FFMs). As a special case, we propose the velvet feedback matrix, which can create dense impulse responses at a minimal computational cost. Further, FFMs can be used to emulate the scattering effects of non-specular reflections. We demonstrate the effectiveness of FFMs in terms of echo density and modal distribution.