Chee Siang Leow
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chee Siang Leow has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Hiromitsu Nishizaki
University of Yamanashi
8 shared publications - Hongqing Liu
Chongqing University of Posts and Telecommunications
3 shared publications - Junkang Yang
University of Yamanashi
3 shared publications - Lu Gan
College of Engineering, Design and Physical Sciences, Brunel University of London, London, U.K.
3 shared publications - Hideaki Yajima
University of Yamanashi
2 shared publications - Changgeng Li
Hangzhou Dianzi University
1 shared publication
Latest Publications
- 14 September 2026
- Published by Springer Nature
- p. 623-634
Harmonic and Entropy Neural Modeling with Demographic Fusion for Speech-Based Depression Detection
- 14 September 2026
- Published by Springer Nature
- p. 623-634
Harmonic and Entropy Neural Modeling with Demographic Fusion for Speech-Based Depression Detection
- 7 September 2026
- Published by Springer Nature
- p. 177-194
Difficulty-Aware Curriculum and Token-Weighted Rationale Distillation from LLM for Sequential Recommendation
- 7 September 2026
- Published by Springer Nature
- p. 177-194
Difficulty-Aware Curriculum and Token-Weighted Rationale Distillation from LLM for Sequential Recommendation
- 1 September 2026
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 160 (3) , 2272-2287
A dual-stream spectral network for versatile full-band speech enhancement
Full-band speech restoration aims to recover high-fidelity speech from signals degraded by noise, reverberation, bandwidth limitation, or their combinations. Existing methods are typically designed for a specific type of degradation, which limits their flexibility when multiple distortions coexist. To address this problem, this paper proposes a dual-stream versatile speech enhancement network (DSVN)—a unified task-conditioned system for full-band speech restoration. The model introduces a dual-stream spectral block to jointly model frequency-wise spectral structures and temporal dynamics. A band-refill decoder is further designed to recover missing or degraded high-frequency components. Task-conditioned modulation is introduced to enable the same network to adapt to different objectives. Evaluation results demonstrate that DSVN improves restoration quality under both single- and mixed-degradation scenarios. The model shows clear advantages in non-intrusive quality assessment, spectral distance, and automatic speech recognition evaluation, indicating that the enhanced speech is not only cleaner but also more useful for downstream systems. Additional analysis of internal responses and ablation studies provides evidence that the proposed modules contribute to different aspects of restoration.
- 1 September 2026
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 160 (3) , 2272-2287
A dual-stream spectral network for versatile full-band speech enhancement
Full-band speech restoration aims to recover high-fidelity speech from signals degraded by noise, reverberation, bandwidth limitation, or their combinations. Existing methods are typically designed for a specific type of degradation, which limits their flexibility when multiple distortions coexist. To address this problem, this paper proposes a dual-stream versatile speech enhancement network (DSVN)—a unified task-conditioned system for full-band speech restoration. The model introduces a dual-stream spectral block to jointly model frequency-wise spectral structures and temporal dynamics. A band-refill decoder is further designed to recover missing or degraded high-frequency components. Task-conditioned modulation is introduced to enable the same network to adapt to different objectives. Evaluation results demonstrate that DSVN improves restoration quality under both single- and mixed-degradation scenarios. The model shows clear advantages in non-intrusive quality assessment, spectral distance, and automatic speech recognition evaluation, indicating that the enhanced speech is not only cleaner but also more useful for downstream systems. Additional analysis of internal responses and ablation studies provides evidence that the proposed modules contribute to different aspects of restoration.
- 24 August 2026
- Published by Springer Nature
- p. 3-19
Reference-Free Handwritten Japanese Character Generation via CLIP-Conditioned Diffusion Models
- 24 August 2026
- Published by Springer Nature
- p. 3-19
Reference-Free Handwritten Japanese Character Generation via CLIP-Conditioned Diffusion Models
- Open Access
- 28 July 2026
- Published by Elsevier in Internet of Things
- Vol. 39
Lightweight streaming speech enhancement for AIoT-enabled wearable hearing aids using parallel spiking mamba
- 28 July 2026
- Published by Elsevier in Internet of Things
- Vol. 39
Open AccessLightweight streaming speech enhancement for AIoT-enabled wearable hearing aids using parallel spiking mamba
- Open Access
- 15 July 2025
- Published by American Chemical Society (ACS) in ChemRxiv
ColabReaction: Accelerating Transition State Searches with Machine Learning Potentials on Google Colaboratory
We have developed a rapid and automated transition state (TS) search method for chemical reactions by combining the double-ended method, Direct MaxFlux (DMF), with machine learning (ML) potentials. Compared to conventional quantum mechanical (QM) scan-based approaches, this method achieves approximately two orders of magnitude speedup, typically locating TS structures within several minutes. To promote broad accessibility, this method is implemented on Google Colaboratory (Colab), leveraging its cloud-based GPU environment to eliminate the need for local computational resources. We named this implementation ColabReaction. A modified Panel-based graphical user interface is also provided, allowing users to perform TS searches through a web-based interface without writing code. This platform offers a cost-free, user-friendly solution for reaction pathway exploration and mechanistic analysis, particularly for experimental researchers and students without prior experience in computational chemistry. ColabReaction is open-source and freely available at https://ColabReaction.net and https://github.com/BILAB/ColabReaction.
- 15 July 2025
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessColabReaction: Accelerating Transition State Searches with Machine Learning Potentials on Google Colaboratory
We have developed a rapid and automated transition state (TS) search method for chemical reactions by combining the double-ended method, Direct MaxFlux (DMF), with machine learning (ML) potentials. Compared to conventional quantum mechanical (QM) scan-based approaches, this method achieves approximately two orders of magnitude speedup, typically locating TS structures within several minutes. To promote broad accessibility, this method is implemented on Google Colaboratory (Colab), leveraging its cloud-based GPU environment to eliminate the need for local computational resources. We named this implementation ColabReaction. A modified Panel-based graphical user interface is also provided, allowing users to perform TS searches through a web-based interface without writing code. This platform offers a cost-free, user-friendly solution for reaction pathway exploration and mechanistic analysis, particularly for experimental researchers and students without prior experience in computational chemistry. ColabReaction is open-source and freely available at https://ColabReaction.net and https://github.com/BILAB/ColabReaction.
- Open Access
- 26 May 2025
- Published by Springer Nature in The Visual Computer
- Vol. 41 (14) , 11705-11720
Non-invasive estimation of Shine Muscat grape color and sensory evaluation from standard camera images
This study proposes a non-invasive method to estimate both color and sensory attributes of Shine Muscat grapes from standard camera images. First, we focus on color estimation by integrating a Vision Transformer (ViT) feature extractor with interquartile range (IQR)-based outlier removal. Experimental results show that our approach achieves 97.2% accuracy, significantly outperforming Convolutional Neural Network (CNN) models. This improvement underscores the importance of capturing global contextual information to differentiate subtle color variations in grape ripeness. Second, we address human sensory evaluation by collecting questionnaire responses on 13 attributes (e.g., “Sweetness,” “Overall taste rating”), each rated on a five-point scale. Because these ratings tend to cluster around midrange values (labels “2,” “3,” and “4”), we initially limit the dataset to the extreme labels “1” (“lowest grade”) and “5” (“highest grade”) for binary classification. Three attributes—“Overall color,” “Sweetness,” and “Overall taste rating”—exhibit relatively high classification accuracies of 79.9%, 75.1%, and 75.7%, respectively. By contrast, the other 10 attributes reach only 50%–66%, suggesting that subjective variations and limited visual cues pose significant challenges. Overall, the proposed approach demonstrates the feasibility of an image-based system that integrates color estimation and sensory evaluation to support more objective, data-driven harvest timing decisions for Shine Muscat grapes.
- 26 May 2025
- Published by Springer Nature in The Visual Computer
- Vol. 41 (14) , 11705-11720
Open AccessNon-invasive estimation of Shine Muscat grape color and sensory evaluation from standard camera images
This study proposes a non-invasive method to estimate both color and sensory attributes of Shine Muscat grapes from standard camera images. First, we focus on color estimation by integrating a Vision Transformer (ViT) feature extractor with interquartile range (IQR)-based outlier removal. Experimental results show that our approach achieves 97.2% accuracy, significantly outperforming Convolutional Neural Network (CNN) models. This improvement underscores the importance of capturing global contextual information to differentiate subtle color variations in grape ripeness. Second, we address human sensory evaluation by collecting questionnaire responses on 13 attributes (e.g., “Sweetness,” “Overall taste rating”), each rated on a five-point scale. Because these ratings tend to cluster around midrange values (labels “2,” “3,” and “4”), we initially limit the dataset to the extreme labels “1” (“lowest grade”) and “5” (“highest grade”) for binary classification. Three attributes—“Overall color,” “Sweetness,” and “Overall taste rating”—exhibit relatively high classification accuracies of 79.9%, 75.1%, and 75.7%, respectively. By contrast, the other 10 attributes reach only 50%–66%, suggesting that subjective variations and limited visual cues pose significant challenges. Overall, the proposed approach demonstrates the feasibility of an image-based system that integrates color estimation and sensory evaluation to support more objective, data-driven harvest timing decisions for Shine Muscat grapes.
- Open Access
- 1 January 2025
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
Handwritten Character Image Generation for Effective Data Augmentation
This study introduces data augmentation techniques to enhance training datasets for a Japanese handwritten character classification model, addressing the high cost of collecting extensive handwritten character data. A novel method is proposed to automatically generate a largescale dataset of handwritten characters from a smaller dataset, utilizing a style transformation approach, particularly Adaptive Instance Normalization (AdaIN).Additionally, the study presents an innovative technique to improve character structural information by integrating features from the Contrastive Language-Image Pre-training (CLIP) text encoder. This approach enables the creation of diverse handwritten character images, including Kanji, by merging content and style elements. The effectiveness of our approach is demonstrated by evaluating a handwritten character classification model using an expanded dataset, which includes Japanese hiragana, katakana, and Kanji from the ETL Character Database. The character classification model's macro F1 score improved from 0.9733 with the original dataset to 0.9861 using the augmented dataset by the proposed approach. This result indicated that our proposed character generation model was able to generate new character images that were not included in the original dataset and that they effectively contributed to training the handwritten character classification model.
- 1 January 2025
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
Open AccessHandwritten Character Image Generation for Effective Data Augmentation
This study introduces data augmentation techniques to enhance training datasets for a Japanese handwritten character classification model, addressing the high cost of collecting extensive handwritten character data. A novel method is proposed to automatically generate a largescale dataset of handwritten characters from a smaller dataset, utilizing a style transformation approach, particularly Adaptive Instance Normalization (AdaIN).Additionally, the study presents an innovative technique to improve character structural information by integrating features from the Contrastive Language-Image Pre-training (CLIP) text encoder. This approach enables the creation of diverse handwritten character images, including Kanji, by merging content and style elements. The effectiveness of our approach is demonstrated by evaluating a handwritten character classification model using an expanded dataset, which includes Japanese hiragana, katakana, and Kanji from the ETL Character Database. The character classification model's macro F1 score improved from 0.9733 with the original dataset to 0.9861 using the augmented dataset by the proposed approach. This result indicated that our proposed character generation model was able to generate new character images that were not included in the original dataset and that they effectively contributed to training the handwritten character classification model.