Disentanglement learning for variational autoencoders applied to audio-visual speech enhancement

G Carbajal, J Richter… - 2021 IEEE Workshop on …, 2021 - ieeexplore.ieee.org
Recently, the standard variational autoencoder has been successfully used to learn a
probabilistic prior over speech signals, which is then used to perform speech enhancement …

Automatic Bird Sound Source Separation Based On Passive Acoustic Devices in Wild Environment

J Xie, Y Shi, D Ni, M Milling, S Liu… - IEEE Internet of …, 2024 - ieeexplore.ieee.org
The Internet of Things (IoT)-based passive acoustic monitoring (PAM) has shown great
potential in large-scale remote bird monitoring. However, field recordings often contain …

[PDF][PDF] Speaker-and phone-aware convolutional transformer network for acoustic echo cancellation

C Han, W Tu, Y Yang, J Li, X Li - Proc. Interspeech 2022, 2022 - drive.google.com
Recent studies indicate the effectiveness of deep learning (DL) based methods for acoustic
echo cancellation (AEC) in background noise and nonlinear distortion scenarios. However …

Robust Semi-Supervised Regression for Vehicle Interior Noise Prediction

S Sim, J Bae, SB Kim - IEEE Access, 2023 - ieeexplore.ieee.org
The rapid advancement of artificial intelligence has observed increased application in
predicting vehicle interior noise levels within the automotive industry. However, the …

指定输出通道排序的半监督盲源分离算法

顾昭仪, 卢晶 - 南京大学学报(自然科学版), 2021 - jns.nju.edu.cn
摘要在频域操作的联合盲源分离算法可以有效解决频点间的内部排序问题, 然而对于输出通道的
排序, 即全局排序, 现有的基于频域的联合盲源分离算法仍无法有效确定. 使用基于变分自编码器 …

Role of Speech Separation in Verifying the Speaker Under Degraded Conditions Using EMD and Hilbert Transform

MKP Kumar, R Kumaraswamy - Proceedings of the International …, 2022 - Springer
In this paper, we discuss the role of separating speech signals in verifying the speaker
effectively under conditions like background noise and multi-speaker environment. The key …

[PDF][PDF] A Unified Statistical Approach to Fast and Robust Multichannel Speech Separation and Dereverberation

K Sekiguchi - 2021 - repository.kulib.kyoto-u.ac.jp
This thesis describes a unified statistical approach to joint multichannel source separation
and dereverberation. This technique is useful as a front end of various audio applications …