Search Results - "Timo Gerkmann"

Bayesian Estimation of Clean Speech Spectral Coefficients Given a Priori Knowledge of the Phase by Gerkmann, Timo

“…While most short-time discrete Fourier transform-based single-channel speech enhancement algorithms only modify the noisy spectral amplitude, in recent years…”
Get full text

STFT Phase Reconstruction in Voiced Speech for an Improved Single-Channel Speech Enhancement by Krawczyk, Martin, Gerkmann, Timo

“…The enhancement of speech which is corrupted by noise is commonly performed in the short-time discrete Fourier transform domain. In case only a single…”
Get full text

On MMSE-Based Estimation of Amplitude and Complex Speech Spectral Coefficients Under Phase-Uncertainty by Krawczyk-Becker, Martin, Gerkmann, Timo

“…Among the most commonly used single-channel approaches for the enhancement of noise corrupted speech are Bayesian estimators of clean speech coefficients in…”
Get full text

A neural network-supported two-stage algorithm for lightweight dereverberation on hearing devices by Lemercier, Jean-Marie, Thiemann, Joachim, Koning, Raphael, Gerkmann, Timo

“…A two-stage lightweight online dereverberation algorithm for hearing devices is presented in this paper. The approach combines a multi-channel multi-frame…”
Get full text

Causal Diffusion Models for Generalized Speech Enhancement by Richter, Julius, Welker, Simon, Lemercier, Jean-Marie, Lay, Bunlong, Peer, Tal, Gerkmann, Timo

“…In this work, we present a causal speech enhancement system that is designed to handle different types of corruptions. This paper is an extended version of our…”
Get full text

Front-end technologies for robust ASR in reverberant environments—spectral enhancement-based dereverberation and auditory modulation filterbank features by Xiong, Feifei, Meyer, Bernd T., Moritz, Niko, Rehr, Robert, Anemüller, Jörn, Gerkmann, Timo, Doclo, Simon, Goetze, Stefan

“…This paper presents extended techniques aiming at the improvement of automatic speech recognition (ASR) in single-channel scenarios in the context of the…”
Get full text

A Survey on Probabilistic Models in Human Perception and Machines by Li, Lux, Rehr, Robert, Bruns, Patrick, Gerkmann, Timo, Röder, Brigitte

“…Extracting information from noisy signals is of fundamental importance for both biological and artificial perceptual systems. To provide tractable solutions to…”
Get full text

Noise power estimation based on the probability of speech presence by Gerkmann, T., Hendriks, R. C.

“…In this paper, we analyze the minimum mean square error (MMSE) based spectral noise power estimator [1] and present an improvement. We will show that the MMSE…”
Get full text

A Multi-Phase Gammatone Filterbank for Speech Separation Via Tasnet by Ditter, David, Gerkmann, Timo

“…In this work, we investigate if the learned encoder of the end-to-end convolutional time domain audio separation network (Conv-TasNet) is the key to its recent…”
Get full text

Insights Into Deep Non-Linear Filters for Improved Multi-Channel Speech Enhancement by Tesch, Kristina, Gerkmann, Timo

“…The key advantage of using multiple microphones for speech enhancement is that spatial filtering can be used to complement the tempo-spectral processing. In a…”
Get full text

Multi-Channel Speech Separation Using Spatially Selective Deep Non-Linear Filters by Tesch, Kristina, Gerkmann, Timo

“…In a multi-channel separation task with multiple speakers, we aim to recover all individual speech signals from the mixture. In contrast to single-channel…”
Get full text

Phase-aware deep speech enhancement: It's all about the frame length by Peer, Tal, Gerkmann, Timo

“…Algorithmic latency in speech processing is dominated by the frame length used for Fourier analysis, which in turn limits the achievable performance of…”
Get full text

SNR-Based Features and Diverse Training Data for Robust DNN-Based Speech Enhancement by Rehr, Robert, Gerkmann, Timo

“…In this paper, we address the generalization of deep neural network (DNN) based speech enhancement to unseen noise conditions for the case that training data…”
Get full text

Spatially Selective Deep Non-Linear Filters For Speaker Extraction by Tesch, Kristina, Gerkmann, Timo

“…In a scenario with multiple persons talking simultaneously, the spatial characteristics of the signals are the most distinct feature for extracting the target…”
Get full text

Nonlinear Spatial Filtering in Multichannel Speech Enhancement by Tesch, Kristina, Gerkmann, Timo

“…The majority of multichannel speech enhancement algorithms are two-step procedures that first apply a linear spatial filter, a so-called beamformer, and…”
Get full text

Speech Enhancement and Dereverberation with Diffusion-based Generative Models by Richter, Julius, Welker, Simon, Lemercier, Jean-Marie, Lay, Bunlong, Gerkmann, Timo

“…In this work, we build upon our previous publication and use diffusion-based generative models for speech enhancement. We present a detailed overview of the…”
Get full text

Uncertainty Estimation in Deep Speech Enhancement Using Complex Gaussian Mixture Models by Fang, Huajian, Gerkmann, Timo

“…Single-channel deep speech enhancement approaches often estimate a single multiplicative mask to extract clean speech without a measure of its accuracy…”
Get full text

Unbiased MMSE-Based Noise Power Estimation With Low Complexity and Low Tracking Delay by Gerkmann, T., Hendriks, R. C.

“…Recently, it has been proposed to estimate the noise power spectral density by means of minimum mean-square error (MMSE) optimal estimation. We show that the…”
Get full text

DriftRec: Adapting Diffusion Models to Blind JPEG Restoration by Welker, Simon, Chapman, Henry N., Gerkmann, Timo

“…In this work, we utilize the high-fidelity generation abilities of diffusion models to solve blind JPEG restoration at high compression levels. We propose an…”
Get full text

Distilling Hubert with LSTMs via Decoupled Knowledge Distillation by de Oliveira, Danilo, Gerkmann, Timo

“…Much research effort is being applied to the task of compressing the knowledge of self-supervised models, which are powerful, yet large and memory consuming…”
Get full text