Byeong-Yeol Kim

Byeong-Yeol Kim

Auto-generated profile, based on publicly available information

Byeong-Yeol Kim

42dot Inc., Hyundai Motor Company, Seoul, Republic of Korea

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 October 2023
Verified
Conference Paper
Open access
YJYoungjoon Jang
Youngjoon Jang
KRKyeongha Rho
Kyeongha Rho
JWJongbin Woo
Jongbin Woo
HLHyeongkeun Lee
Hyeongkeun Lee
JPJihwan Park
Jihwan Park
JCJoon Son Chung
Joon Son Chung

9 January 2023
Verified
Conference Paper
Closed access
HKHyung Yong Kim
Hyung Yong Kim
BKByeong-Yeol Kim
Byeong-Yeol Kim
SYSeung Woo Yoo
Seung Woo Yoo
YLYoushin Lim
Youshin Lim
YLYunkyu Lim
Yunkyu Lim
HLHanbin Lee
Hanbin Lee
Pre-training of self-supervised learning (SSL) generally shows a good performance on various speech processing tasks. However, this pre-training scheme may lead to a sub-optimal solution for fine-tuning a specific task, such as automatic speech recognition (ASR). In order to provide a more optimal pre-trained model for ASR, we introduce an ASR-Specific hidden-unit BERT with self-training, namely ASBERT. Motivated by self-training, we extract linguistic-related pseudo labels from the fine-tuned model, and these labels are used in the next pre-training procedure. Experimental results on LibriSpeech test-clean and test-other datasets show that ASBERT without language model (LM) outperforms the conventional SSL and self-training model, achieving a 6.3/2.0% and 15.4/13.2% relatively word error rate reduction (RERR). Moreover, without using pseudo-transcription, ASBERT yields comparable performance to the conventional self-training method.

1 November 2022 in arXiv
Verified
Preprint
Open access
PDF file
JJJaemin Jung
Jaemin Jung
YKYoukyum Kim
Youkyum Kim
JPJihwan Park
Jihwan Park
YLYoushin Lim
Youshin Lim
BKByeong-Yeol Kim
Byeong-Yeol Kim
JCJoon Son Chung
Joon Son Chung
The goal of this work is to detect new spoken terms defined by users. While most previous works address Keyword Spotting (KWS) as a closed-set classification problem, this limits their transferability to unseen terms. The ability to define custom keywords has advantages in terms of user experience. In this paper, we propose a metric learning-based training strategy for user-defined keyword spotting. In particular, we make the following contributions: (1) we construct a large-scale keyword dataset with an existing speech corpus and propose a filtering method to remove data that degrade model training; (2) we propose a metric learning-based two-stage training strategy, and demonstrate that the proposed method improves the performance on the user-defined keyword spotting task by enriching their representations; (3) to facilitate the fair comparison in the user-defined KWS field, we propose unified evaluation protocol and metrics. Our proposed system does not require an incremental training on the user-defined keywords, and outperforms previous works by a significant margin on the Google Speech Commands dataset using the proposed as well as the existing metrics.