On loss functions for supervised monaural time-domain speech enhancement

M Kolbæk, ZH Tan, SH Jensen… - IEEE/ACM Transactions …, 2020 - ieeexplore.ieee.org
Many deep learning-based speech enhancement algorithms are designed to minimize the
mean-square error (MSE) in some transform domain between a predicted and a target …

Real-time denoising and dereverberation wtih tiny recurrent u-net

HS Choi, S Park, JH Lee, H Heo… - ICASSP 2021-2021 …, 2021 - ieeexplore.ieee.org
Modern deep learning-based models have seen outstanding performance improvement with
speech enhancement tasks. The number of parameters of state-of-the-art models, however …

[HTML][HTML] A Survey on Low-Latency DNN-Based Speech Enhancement

S Drgas - Sensors, 2023 - mdpi.com
This paper presents recent advances in low-latency, single-channel, deep neural network-
based speech enhancement systems. The sources of latency and their acceptable values in …

UNetGAN: A robust speech enhancement approach in time domain for extremely low signal-to-noise ratio condition

X Hao, X Su, Z Wang, H Zhang - arXiv preprint arXiv:2010.15521, 2020 - arxiv.org
Speech enhancement at extremely low signal-to-noise ratio (SNR) condition is a very
challenging problem and rarely investigated in previous works. This paper proposes a …

Real-time single-channel speech enhancement based on causal attention mechanism

J Fan, J Yang, X Zhang, Y Yao - Applied Acoustics, 2022 - Elsevier
To achieve real-time single-channel speech enhancement, ie, enhancing with no or low
latency, this paper proposes a causal speech enhancement model with an attention …

[PDF][PDF] Spectro-Temporal SubNet for Real-Time Monaural Speech Denoising and Dereverberation.

F Xiong, W Chen, P Wang, X Li, J Feng - Interspeech, 2022 - researchgate.net
This paper presents an improved subband neural network applied to joint speech denoising
and dereverberation for online single-channel scenarios. Preserving the advantages of …

Combining multi-perspective attention mechanism with convolutional networks for monaural speech enhancement

T Lan, Y Lyu, W Ye, G Hui, Z Xu, Q Liu - IEEE Access, 2020 - ieeexplore.ieee.org
The redundant convolutional encoder-decoder network has been proven useful in speech
enhancement tasks. This network can capture the localized time-frequency details of speech …

[PDF][PDF] Low-Delay Speech Enhancement Using Perceptually Motivated Target and Loss.

X Zhang, X Ren, X Zheng, L Chen, C Zhang, L Guo… - Interspeech, 2021 - isca-archive.org
Speech enhancement approaches based on deep neural network have outperformed the
traditional signal processing methods. This paper presents a low-delay speech …

Speech enhancement using U-nets with wide-context units

T Grzywalski, S Drgas - Multimedia Tools and Applications, 2022 - Springer
In this article a new neural network for speech enhancement is proposed where single-
channel noisy speech is processed in order to improve its intelligibility and quality. It is …

[PDF][PDF] Monoaural Speech Enhancement Using a Nested U-Net with Two-Level Skip Connections.

S Hwang, Y Park, S Park - INTERSPEECH, 2022 - isca-archive.org
Capturing the contextual information in multi-scale is known to be beneficial for improving
the performance of DNN-based speech enhancement (SE) models. This paper proposes a …