An overview of noise-robust automatic speech recognition
New waves of consumer-centric applications, such as voice search and voice interaction
with mobile devices and home entertainment systems, increasingly require automatic …
with mobile devices and home entertainment systems, increasingly require automatic …
Sixty years of frequency-domain monaural speech enhancement: From traditional to deep learning methods
Frequency-domain monaural speech enhancement has been extensively studied for over
60 years, and a great number of methods have been proposed and applied to many …
60 years, and a great number of methods have been proposed and applied to many …
Unbiased MMSE-based noise power estimation with low complexity and low tracking delay
T Gerkmann, RC Hendriks - IEEE Transactions on Audio …, 2011 - ieeexplore.ieee.org
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 resulting …
minimum mean-square error (MMSE) optimal estimation. We show that the resulting …
[图书][B] DFT-domain based single-microphone noise reduction for speech enhancement
As speech processing devices like mobile phones, voice controlled devices, and hearing
aids have increased in popularity, people expect them to work anywhere and at any time …
aids have increased in popularity, people expect them to work anywhere and at any time …
Noise power estimation based on the probability of speech presence
T Gerkmann, RC Hendriks - … of Signal Processing to Audio and …, 2011 - ieeexplore.ieee.org
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 based …
power estimator [1] and present an improvement. We will show that the MMSE based …
Recent developments in speech enhancement in the short-time Fourier transform domain
In this paper, we present an overview on the topic of noise reduction in the short-time Fourier
transform (STFT) domain. First, we briefly review the conventional literature in the single-and …
transform (STFT) domain. First, we briefly review the conventional literature in the single-and …
A new linear MMSE filter for single channel speech enhancement based on nonnegative matrix factorization
N Mohammadiha, T Gerkmann… - 2011 IEEE workshop on …, 2011 - ieeexplore.ieee.org
In this paper, a linear MMSE filter is derived for single-channel speech enhancement which
is based on Nonnegative Matrix Factorization (NMF). Assuming an additive model for the …
is based on Nonnegative Matrix Factorization (NMF). Assuming an additive model for the …
A novel fast nonstationary noise tracking approach based on MMSE spectral power estimator
Estimating the noise power spectral density (PSD) from the corrupted speech signal is an
essential component for speech enhancement algorithms. In this paper, a novel noise PSD …
essential component for speech enhancement algorithms. In this paper, a novel noise PSD …
An Analysis of Traditional Noise Power Spectral Density Estimators Based on the Gaussian Stochastic Volatility Model
JK Nielsen, MG Christensen… - IEEE/ACM Transactions …, 2023 - ieeexplore.ieee.org
Many single-and multi-channel speech enhancement techniques, old and new, rely in one
way or another on estimates of the noise power spectral density (PSD). For example, the …
way or another on estimates of the noise power spectral density (PSD). For example, the …
A cepstrum-based preprocessing and postprocessing for speech enhancement in adverse environments
X Hu, S Wang, C Zheng, X Li - Applied acoustics, 2013 - Elsevier
This paper proposes a cepstrum-based preprocessing and postprocessing algorithm for
single-channel speech enhancement. The cepstrum-based preprocessing scheme is …
single-channel speech enhancement. The cepstrum-based preprocessing scheme is …