Oh-Wook Kwon

Prof. Oh-Wook Kwon

orcid

Research Focus & Metrics

Academic Career Length

31Years (1996-Present)

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

Representative Publications

View All Publications
26 June 2019 in EURASIP Journal on Audio, Speech, and Music Processing
Verified
Journal Article
Open access
PDF file
BJByeong-Yong Jang
Byeong-Yong Jang
WHWoon-Haeng Heo
Woon-Haeng Heo
JKJung-Hyun Kim
Jung-Hyun Kim
OKOh-Wook Kwon
Oh-Wook Kwon
We propose a new method for music detection from broadcasting contents using the convolutional neural networks with a Mel-scale kernel. In this detection task, music segments should be annotated from the broadcast data, where music, speech, and noise are mixed. The convolutional neural network is composed of a convolutional layer with kernel that is trained to extract robust features. The Mel-scale changes the kernel size, and the backpropagation algorithm trains the kernel shape. We used 52 h of mixed broadcast data (25 h of music) to train the convolutional network and 24 h of collected broadcast data (ratio of music of 50–76%) for testing. The test data consisted of various genres (drama, documentary, news, kids, reality, and so on) that are broadcast in British English, Spanish, and Korean languages. The proposed method consistently showed better performance in all the three languages than the baseline system, and the F-score ranged from 86.5% for British data to 95.9% for Korean drama data. Our music detection system takes about 28 s to process a 1-min signal using only one CPU with 4 cores.

1 February 2020 in IEICE Transactions on Information and Systems
Verified
Journal Article
Open access
PDF file
JBJeong-Uk Bang
Jeong-Uk Bang
MCMu-Yeol Choi
Mu-Yeol Choi
SKSanghun Kim
Sanghun Kim
OKOh-Wook Kwon
Oh-Wook Kwon
Japan's largest platform for academic e-journals: J-STAGE is a full text database for reviewed academic papers published by Japanese societies

3 March 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
WHWoon-Haeng Heo
Woon-Haeng Heo
HKHyemi Kim
Hyemi Kim
OKOh-Wook Kwon
Oh-Wook Kwon
We propose a source separation architecture using dilated time-frequency DenseNet for background music identification of broadcast content. We apply source separation techniques to the mixed signals of music and speech. For the source separation purpose, we propose a new architecture to add a time-frequency dilated convolution to the conventional DenseNet in order to effectively increase the receptive field in the source separation scheme. In addition, we apply different convolutions to each frequency band of the spectrogram in order to reflect the different frequency characteristics of the low- and high-frequency bands. To verify the performance of the proposed architecture, we perform singing-voice separation and music-identification experiments. As a result, we confirm that the proposed architecture produces the best performance in both experiments because it uses the dilated convolution to reflect wide contextual information.

1 September 2011
Verified
Conference Paper
Closed access
YLYun-Kyung Lee
Yun-Kyung Lee
OKOh-Wook Kwon
Oh-Wook Kwon
HSHyun Soon Shin
Hyun Soon Shin
JJJun Jo
Jun Jo
YLYongkwi Lee
Yongkwi Lee
In this paper, we address the problem of noise reduction of photoplethysmography (PPG) signals acquired from an PPG array sensor. The previous noise reduction approaches assumed that the noise sources are stationary. However, in real environments PPG signals often get corrupted by nonstationary movement noise. To reduce such noise, we propose to estimate the desired signal from corrupted signals by using a particle filter. In computer experiments using real PPG signals acquired from a wristwatch-type PPG array sensor, the proposed algorithm is shown to effectively reduce the movement noise and improve emotion recognition accuracy absolutely by 12.7 % and 10.9 % in the situations where users move arms and walk on a road, respectively, compared with the conventional normalized least-mean-square (NLMS)-based algorithm. The output signal-to-noise ratio (SNR) is also improved by 4.5 dB on average in the same situations.

1 January 1996 in IEEE Transactions on Speech and Audio Processing
Verified
Journal Article
Closed access
YPY.K. Park
Y.K. Park
CUC.K. Un
C.K. Un
OKOh-Wook Kwon
Oh-Wook Kwon

1 November 2010 in IEEE Transactions on Consumer Electronics
Verified
Journal Article
Closed access
YLYun-Kyung Lee
Yun-Kyung Lee
ILIn Sung Lee
In Sung Lee
OKOh-Wook Kwon
Oh-Wook Kwon
This paper addresses the problem of single-channel speech separation to extract and enhance the desired speech signal from mixed speech signals. We propose a new speech separation algorithm by utilizing both magnitude and phase information, which can be applied to multimedia mobile communication and navigation systems. Conventionally, phase information has been neglected in speech signal processing. However, in the proposed method, we formulate a probabilistic phase-based speech estimator based on zero-phase models to improve the speech separation performance. In the speech separation experiments, the proposed method is shown to improve speaker-to-interference ratio (SIR) by 2.2 dB compared to the system using magnitude models only. When only phase-based speech estimator is used for speech separation, the SIR was improved by 0.8 dB. This result justify that the proposed phase-based speech estimator achieves significant SIR improvement compared with the previous magnitude-based method.