no code implementations • 31 Aug 2023 • Seunghan Yang, Byeonggeun Kim, Kyuhong Shim, Simyung Chang
Few-shot keyword spotting (FS-KWS) models usually require large-scale annotated datasets to generalize to unseen target keywords.
no code implementations • 26 Feb 2023 • Byeonggeun Kim, Jun-Tae Lee, Seunghan Yang, Simyung Chang
Efficient transfer learning involves utilizing a pre-trained model trained on a larger dataset and repurposing it for downstream tasks with the aim of maximizing the reuse of the pre-trained model.
no code implementations • 10 Feb 2023 • Hyesu Lim, Byeonggeun Kim, Jaegul Choo, Sungha Choi
In this paper, we identify that CBN and TBN are in a trade-off relationship and present a new test-time normalization (TTN) method that interpolates the statistics by adjusting the importance between CBN and TBN according to the domain-shift sensitivity of each BN layer.
no code implementations • 28 Jun 2022 • Byeonggeun Kim, Seunghan Yang, Jangho Kim, Simyung Chang
The goal of the task is to design an audio scene classification system for device-imbalanced datasets under the constraints of model complexity.
no code implementations • 28 Jun 2022 • Seunghan Yang, Byeonggeun Kim, Inseop Chung, Simyung Chang
We design two personalized KWS tasks; (1) Target user Biased KWS (TB-KWS) and (2) Target user Only KWS (TO-KWS).
no code implementations • 28 Jun 2022 • Byeonggeun Kim, Seunghan Yang, Inseop Chung, Simyung Chang
We also verify our method on a standard benchmark, miniImageNet, and D-ProtoNets shows the state-of-the-art open-set detection rate in FSOSR.
no code implementations • 24 Jun 2022 • Byeonggeun Kim, Seunghan Yang, Jangho Kim, Hyunsin Park, JunTae Lee, Simyung Chang
While using two-dimensional convolutional neural networks (2D-CNNs) in image processing, it is possible to manipulate domain information using channel statistics, and instance normalization has been a promising way to get domain-invariant features.
no code implementations • 12 Nov 2021 • Byeonggeun Kim, Seunghan Yang, Jangho Kim, Simyung Chang
Moreover, we introduce an efficient architecture, BC-ResNet-ASC, a modified version of the baseline architecture with a limited receptive field.
no code implementations • 29 Sep 2021 • Byeonggeun Kim, Seunghan Yang, Jangho Kim, Hyunsin Park, Jun-Tae Lee, Simyung Chang
While using two-dimensional convolutional neural networks (2D-CNNs) in image processing, it is possible to manipulate domain information using channel statistics, and instance normalization has been a promising way to get domain-invariant features.
4 code implementations • 8 Jun 2021 • Byeonggeun Kim, Simyung Chang, Jinkyu Lee, Dooyong Sung
We present a broadcasted residual learning method to achieve high accuracy with small model size and computational load.
Ranked #2 on Keyword Spotting on Google Speech Commands
no code implementations • 11 Oct 2019 • Byeonggeun Kim, Mingu Lee, Jinkyu Lee, Yeonseok Kim, Kyuwoong Hwang
A keyword spotting (KWS) system determines the existence of, usually predefined, keyword in a continuous speech stream.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2
no code implementations • 10 Oct 2019 • Mingu Lee, Jinkyu Lee, Hye Jin Jang, Byeonggeun Kim, Wonil Chang, Kyuwoong Hwang
Augmenting regularization terms which penalize positional and contextual non-orthogonality between the attention heads encourages to output different representations from separate subsequences, which in turn enables leveraging structured information without explicit sequence models such as hidden Markov models.