1 code implementation • 15 Dec 2022 • Dongheon Lee, Jung-Woo Choi
In this study, we propose a dense frequency-time attentive network (DeFT-AN) for multichannel speech enhancement.
Ranked #1 on Speech Enhancement on spatialized DNS challenge
1 code implementation • 18 Apr 2024 • Dongheon Lee, Seokju Yun, Youngmin Ro
As a result, we introduce Partial Large Kernel CNNs for Efficient Super-Resolution (PLKSR), which achieves state-of-the-art performance on four datasets at a scale of $\times$4, with reductions of 68. 1\% in latency and 80. 2\% in maximum GPU memory occupancy compared to SRFormer-light.
no code implementations • 8 Nov 2021 • Dongheon Lee, Seongrae Kim, Jung-Woo Choi
In this study, we propose an end-to-end time-domain speech enhancement network that can facilitate the use of inter-channel relationships at individual layers of a DNN.
Ranked #1 on Speech Enhancement on CHiME-3
no code implementations • 18 Mar 2022 • Dongheon Lee, Gyuyeol Kong, Jang-Won Lee, Sooyong Choi
In this study, a novel coding scheme called highdensity coding based on high-density codebooks using a genetic local search algorithm is proposed.
no code implementations • 30 Aug 2023 • Dongheon Lee, Jung-Woo Choi
In the proposed split dense blocks extracting spatial features, a pair of subgroups is sequentially concatenated and processed by convolution layers to effectively reduce the computational complexity and memory usage.
no code implementations • 29 Jan 2024 • Dongheon Lee, Seungmyong Jeong, Youngmin Ro
Numerical models have long been used to understand geoscientific phenomena, including tidal currents, crucial for renewable energy production and coastal engineering.