The time-domain speech separation methods adopting deep learning have obtained impressive performance. However, the computational complexity, model size, and …
Y Luo, N Mesgarani - IEEE/ACM transactions on audio, speech …, 2019 - ieeexplore.ieee.org
Single-channel, speaker-independent speech separation methods have recently seen great progress. However, the accuracy, latency, and computational cost of such methods remain …
W Tong, J Zhu, J Chen, Z Wu, S Kang… - ICASSP 2023-2023 …, 2023 - ieeexplore.ieee.org
Deep learning-based methods have made significant achievements in speech separation. Especially the time-domain separation methods have achieved the best performance in …
GP Yang, CI Tuan, HY Lee, L Lee - arXiv preprint arXiv:1904.07845, 2019 - arxiv.org
Speech separation has been very successful with deep learning techniques. Substantial effort has been reported based on approaches over spectrogram, which is well known as the …
C Quan, X Li - arXiv preprint arXiv:2204.04464, 2022 - arxiv.org
This work proposes a multichannel speech separation method with narrow-band Conformer (named NBC). The network is trained to learn to automatically exploit narrow-band speech …
Y Zhao, C Luo, ZJ Zha, W Zeng - Proceedings of the Twenty-Ninth …, 2021 - ijcai.org
In this paper, we introduce Transformer to the timedomain methods for single-channel speech separation. Transformer has the potential to boost speech separation performance …
K Wang, H Huang, Y Hu, Z Huang, S Li - Interspeech, 2021 - isca-archive.org
Traditional single channel speech separation in the timefrequency (TF) domain often faces the problem of phase reconstruction. Due to the fact that the real-valued network is not …
L Yang, W Liu, W Wang - ICASSP 2022-2022 IEEE …, 2022 - ieeexplore.ieee.org
Speech separation has been very successful with deep learning techniques. In this paper, we propose time-frequency (TF) domain path scanning network (TFPSNet) for speech …
MWY Lam, J Wang, D Su, D Yu - ICASSP 2021-2021 IEEE …, 2021 - ieeexplore.ieee.org
One of the leading single-channel speech separation (SS) models is based on a TasNet with a dual-path segmentation technique, where the size of each segment remains …