Publications | ISIPLab

2024

CATAD: Conformer-Based Adversarial Training with Adaptive Diffusion for Bone-Conducted Speech Enhancement (submitted)

Jian Zhou, Zhiqiang Duan, Cunhang Fan, and 2 more authors

The 14th International Symposium on Chinese Spoken Language Processing (ISCSLP 2024), 2024

@article{dzq2024,
  author = {Zhou, Jian and Duan, Zhiqiang and Fan, Cunhang and Tao, Liang and Lv, Zhao},
  year = {2024},
  title = {CATAD: Conformer-Based Adversarial Training with Adaptive Diffusion for Bone-Conducted Speech Enhancement (submitted)},
  journal = {The 14th International Symposium on Chinese Spoken Language Processing (ISCSLP 2024)},
}

Region Attention Fused Semi-Supervised Cross-Domain Facial Expression Recognition (submitted)

Jian Zhou, Jia Zhou, Wenming Zheng, and 2 more authors

IEEE transactions on Affective Computing, 2024

Bib

@article{zj2024,
  author = {Zhou, Jian and Zhou, Jia and Zheng, Wenming and Tao, Liang and Kwan, Hon Keung},
  year = {2024},
  title = {Region Attention Fused Semi-Supervised Cross-Domain Facial Expression Recognition (submitted)},
  journal = {IEEE transactions on Affective Computing},
}

THGCN:Temporal Hypergraph Convolutional Network for Subject Independent EEG Emotion Recognition (Accepted, CCF C)

Jian Zhou, Jian Xu, Liang Tao, and 4 more authors

ICONIP2024 (The 31st International Conference on Neural Information Processing), 2024

Bib

@article{xj2024,
  author = {Zhou, Jian and Xu, Jian and Tao, Liang and Wang, Huabin and Fan, Cunhang and Lv, Zhao and Lu, Cheng},
  year = {2024},
  title = {THGCN:Temporal Hypergraph Convolutional Network for Subject Independent EEG Emotion Recognition (Accepted, CCF C)},
  journal = {ICONIP2024 (The 31st International Conference on Neural Information Processing)},
}

Multi-scale 3D-CRU for EEG emotion recognition

Hao Dong, Jian Zhou, Cunhang Fan, and 3 more authors

Biomedical Physics & Engineering Express, 2024

Bib HTML

@article{dh2024,
  author = {Dong, Hao and Zhou, Jian and Fan, Cunhang and Zheng, Wenming and Tao, Liang and Kwan, Hon Keung},
  year = {2024},
  title = {Multi-scale 3D-CRU for EEG emotion recognition},
  volume = {10},
  journal = {Biomedical Physics & Engineering Express},
  doi = {10.1088/2057-1976/ad43f1},
}

Multi-granularity Prosodic Speech Synthesis with Grammar Information (submitted)

Jian Zhou, Jiahui Liu, Cunhang Fan, and 4 more authors

Neural Networks, 2024

Bib

@article{ljh2024,
  author = {Zhou, Jian and Liu, Jiahui and Fan, Cunhang and Zheng, Wenming and Lv, Zhao and Tao, Liang and Kwan, Hon Keung},
  year = {2024},
  title = {Multi-granularity Prosodic Speech Synthesis with Grammar Information (submitted)},
  journal = {Neural Networks},
}

Dynamic Ensemble Teacher-Student Distillation Framework for Light-weight Fake Audio Detection

Jun Xue, Cunhang Fan, Jiangyan Yi, and 2 more authors

IEEE Signal Processing Letters, 2024

Bib HTML

@article{20240805,
  author = {Xue, Jun and Fan, Cunhang and Yi, Jiangyan and Zhou, Jian and Lv, Zhao},
  year = {2024},
  title = {Dynamic Ensemble Teacher-Student Distillation Framework for Light-weight Fake Audio Detection},
  volume = {},
  journal = {IEEE Signal Processing Letters},
  doi = {10.1109/LSP.2024.3431936},
}

Multi-Level Information Aggregation Based Graph Attention Networks Towards Fake Speech Detection

Jian Zhou, Yong Li, Cunhang Fan, and 2 more authors

IEEE Signal Processing Letters, 2024

Bib HTML

@article{ly2024,
  author = {Zhou, Jian and Li, Yong and Fan, Cunhang and Tao, Liang and Kwan, Hon Keung},
  year = {2024},
  title = {Multi-Level Information Aggregation Based Graph Attention Networks Towards Fake Speech Detection},
  volume = {31},
  journal = {IEEE Signal Processing Letters},
  doi = {10.1109/LSP.2024.3408676},
}

Multi-Scale Feature Fusion Transformer Network for End-to-End Single Channel Speech Separation (Submitted)

Jian Zhou, Yinhao Xu, Cunhang Fan, and 3 more authors

Circuits, Systems, and Signal Processing, 2024

Bib

@article{cll2022,
  author = {Zhou, Jian and Xu, Yinhao and Fan, Cunhang and Tao, Liang and Lv, Zhao and Kwan, Hon Keung},
  year = {2024},
  title = {Multi-Scale Feature Fusion Transformer Network for End-to-End Single Channel Speech Separation (Submitted)},
  journal = {Circuits, Systems, and Signal Processing},
}

Enhancing Bone-Conducted Speech with Spectrum Similarity Metric in Adversarial Learning (Under Revision R3)

Yan Pan, Jian Zhou, Huabin Wang, and 3 more authors

Speech Communication, 2024

Bib

@article{py2023,
  author = {Pan, Yan and Zhou, Jian and Wang, Huabin and Zheng, Wenming and Tao, Liang and Kwan, Hon},
  year = {2024},
  title = {Enhancing Bone-Conducted Speech with Spectrum Similarity Metric in Adversarial Learning (Under Revision R3)},
  journal = {Speech Communication},
}

2023

AASADI:Audio Anti-Spoofing Using Aggregated Direct and Indirect Nodes Graph Attention Networks (Submitted)

Yong Li, Jian Zhou, Huabin Wang, and 2 more authors

IEEE Signal Processing Letters, 2023

Bib

@article{ly2023,
  author = {Li, Yong and Zhou, Jian and Wang, Huabin and Tao, Liang and Kwan, Hon},
  year = {2023},
  title = {AASADI:Audio Anti-Spoofing Using Aggregated Direct and Indirect Nodes Graph Attention Networks (Submitted)},
  journal = {IEEE Signal Processing Letters},
}

Controllable Multi-Speaker Emotional Speech Synthesis With Generalization Module (Accepted)

Junjie Zheng, Jian Zhou, Wenming Zheng, and 2 more authors

IEEE Transaction on Affective Computing, 2023

Bib

@article{jj2023,
  author = {Zheng, Junjie and Zhou, Jian and Zheng, Wenming and Tao, Liang and Kwan, Hon},
  year = {2023},
  title = {Controllable Multi-Speaker Emotional Speech Synthesis With Generalization Module (Accepted)},
  journal = {IEEE Transaction on Affective Computing},
}

A Novel Attention-Guided Generative Adversarial Network for Whisper-to-Normal Speech Conversion

Teng Gao, Qing Pan, Jian Zhou, and 3 more authors

Cognitive Computation, Jan 2023

Bib HTML

@article{20230115,
  author = {Gao, Teng and Pan, Qing and Zhou, Jian and Wang, Huabin and Tao, Liang and Kwan, Hon},
  year = {2023},
  month = jan,
  title = {A Novel Attention-Guided Generative Adversarial Network for Whisper-to-Normal Speech Conversion},
  volume = {15},
  journal = {Cognitive Computation},
  doi = {10.1007/s12559-023-10108-9},
}

2022

Replay Attack Detection Based on Raw Waveform and Sinc Filter (Accepted)

Rui Jia, Mingzhe Xia, Cunhang Fan, and 3 more authors

National Conference on Man-Machine Speech Communication，NCMMSC2023, Jan 2022

Bib

@article{jr2022,
  author = {Jia, Rui and Xia, Mingzhe and Fan, Cunhang and Wang, Huabing and Tao, Liang and Zhou, Jian},
  year = {2022},
  title = {Replay Attack Detection Based on Raw Waveform and Sinc Filter (Accepted)},
  journal = {National Conference on Man-Machine Speech Communication，NCMMSC2023},
}

CycleGAN with Dual Adversarial Loss for Bone-Conducted Speech Enhancement

Qing Pan, Yan Pan, Jian Zhou, and 3 more authors

In TENCON 2022 - 2022 IEEE Region 10 Conference (TENCON), Jan 2022

Bib HTML

@inproceedings{9977942,
  author = {Pan, Qing and Pan, Yan and Zhou, Jian and Wang, Huabin and Tao, Liang and Kwan, Hon Keung},
  booktitle = {TENCON 2022 - 2022 IEEE Region 10 Conference (TENCON)},
  title = {CycleGAN with Dual Adversarial Loss for Bone-Conducted Speech Enhancement},
  year = {2022},
  pages = {1-4},
  doi = {10.1109/TENCON55691.2022.9977942},
}

SETransformer: Speech Enhancement Transformer

Weiwei Yu, Jian Zhou, HuaBin Wang, and 1 more author

Cognitive Computation, May 2022

Bib HTML

@article{20220514001,
  author = {Yu, Weiwei and Zhou, Jian and Wang, HuaBin and Tao, Liang},
  year = {2022},
  month = may,
  title = {SETransformer: Speech Enhancement Transformer},
  volume = {14},
  journal = {Cognitive Computation},
  doi = {10.1007/s12559-020-09817-2},
}

Multistage Model for Robust Face Alignment Using Deep Neural Networks

Huabin Wang, Rui Cheng, Jian Zhou, and 2 more authors

Cognitive Computation, May 2022

Bib HTML

@article{20220514,
  author = {Wang, Huabin and Cheng, Rui and Zhou, Jian and Tao, Liang and Kwan, Hon},
  year = {2022},
  month = may,
  title = {Multistage Model for Robust Face Alignment Using Deep Neural Networks},
  volume = {14},
  journal = {Cognitive Computation},
  doi = {10.1007/s12559-021-09846-5},
}

2020

利用深度卷积神经网络将耳语转换为正常语音

连海伦, 周健, 胡雨婷, and 1 more author

声学学报, May 2020

Bib HTML

@article{20200117,
  title = {利用深度卷积神经网络将耳语转换为正常语音},
  journal = {声学学报},
  volume = {45},
  number = {1},
  pages = {137-144},
  year = {2020},
  issn = {0371-0025},
  doi = {10.15949/j.cnki.0371-0025.2020.01.017},
  author = {连海伦 and 周健 and 胡雨婷 and 郑文明},
}

2019

Whisper to Normal Speech Conversion Using Sequence-to-Sequence Mapping Model With Auditory Attention

Hailun Lian, Yuting Hu, Weiwei Yu, and 2 more authors

IEEE Access, May 2019

Bib HTML

@article{8835014,
  author = {Lian, Hailun and Hu, Yuting and Yu, Weiwei and Zhou, Jian and Zheng, Wenming},
  journal = {IEEE Access},
  title = {Whisper to Normal Speech Conversion Using Sequence-to-Sequence Mapping Model With Auditory Attention},
  year = {2019},
  volume = {7},
  pages = {130495-130504},
  doi = {10.1109/ACCESS.2019.2940700},
}

An End to End Method of Whisper Enhancement

Yan Huang, HaiLun Lian, Jian Zhou, and 2 more authors

In 2019 IEEE 2nd International Conference on Information Communication and Signal Processing (ICICSP), May 2019

Bib HTML

@inproceedings{8958563,
  author = {Huang, Yan and Lian, HaiLun and Zhou, Jian and Wang, HuaBin and Tao, Liang},
  booktitle = {2019 IEEE 2nd International Conference on Information Communication and Signal Processing (ICICSP)},
  title = {An End to End Method of Whisper Enhancement},
  year = {2019},
  pages = {246-250},
  doi = {10.1109/ICICSP48821.2019.8958563},
}

Multimodal Voice Conversion Under Adverse Environment Using a Deep Convolutional Neural Network

Jian Zhou, Yuting Hu, Hailun Lian, and 3 more authors

IEEE Access, May 2019

Bib HTML

@article{8913498,
  author = {Zhou, Jian and Hu, Yuting and Lian, Hailun and Wang, Huabin and Tao, Liang and Kwan, Hon Keung},
  journal = {IEEE Access},
  title = {Multimodal Voice Conversion Under Adverse Environment Using a Deep Convolutional Neural Network},
  year = {2019},
  volume = {7},
  pages = {170878-170887},
  doi = {10.1109/ACCESS.2019.2955982},
}

2018

采用L_1/2稀疏约束的梅尔倒谱系数语音重建方法

周健, 刘荣敏, 窦云峰, and 2 more authors

声学学报, May 2018

Bib HTML

@article{20180614,
  title = {采用L<sub>1/2</sub>稀疏约束的梅尔倒谱系数语音重建方法},
  journal = {声学学报},
  volume = {43},
  number = {6},
  pages = {991-999},
  year = {2018},
  issn = {0371-0025},
  doi = {10.15949/j.cnki.0371-0025.2018.06.015},
  author = {周健 and 刘荣敏 and 窦云峰 and 路成 and 陶亮},
}

采用低维特征映射的耳语音向正常音转换

周健, 窦云峰, 刘荣敏, and 2 more authors

声学学报, May 2018

Bib HTML

@article{20180517,
  title = {采用低维特征映射的耳语音向正常音转换},
  journal = {声学学报},
  volume = {43},
  number = {5},
  pages = {855-863},
  year = {2018},
  issn = {0371-0025},
  doi = {10.15949/j.cnki.0371-0025.2018.05.017},
  author = {周健 and 窦云峰 and 刘荣敏 and 王华彬 and 陶亮},
}

2017

A Sparse Analysis Window for Discrete Gabor Transform

Jian Zhou, Xianyong Fang, and Liang Tao

Circuits Syst. Signal Process., Oct 2017

Abs Bib HTML

Here, we analyze sparse analysis windows for complex-valued and real-valued discrete Gabor transforms. A number of theories presented in this study indicate that the existence and uniqueness of a sparse analysis window for a given synthesis window depend to a great extent on the sparsity of the synthesis window. Specifically, the upper and lower bounds on the analysis window sparsity are obtained for synthesis windows of different sparsities and analyzed in detail. Particularly, if the sparsity of a given synthesis window with connected support is equal to the discrete Gabor transform time-shift parameter, the sparsest analysis window is unique. In addition, we propose an algorithm that searches for a sparse analysis window. Compared with existing algorithms, a sparse analysis window with fewer nonzero elements and/or smaller reconstruction errors can be obtained with the proposed algorithm. A sequence of experimental results reveals its superior effectiveness. In addition, a better Gabor representation with high concentration and time—frequency resolution is obtained with the sparse analysis window.
@article{20173610, author = {Zhou, Jian and Fang, Xianyong and Tao, Liang}, title = {A Sparse Analysis Window for Discrete Gabor Transform}, year = {2017}, issue_date = {October 2017}, publisher = {Birkhauser Boston Inc.}, address = {USA}, volume = {36}, number = {10}, issn = {0278-081X}, url = {https://doi.org/10.1007/s00034-017-0510-0}, doi = {10.1007/s00034-017-0510-0}, journal = {Circuits Syst. Signal Process.}, month = oct, pages = {4161-4180}, numpages = {20}, keywords = {Gabor time---frequency analysis, Sparse analysis window, Synthesis window}, }

正交投影非负矩阵的交替方向乘子分解方法

周健, 窦云峰, 刘荣敏, and 2 more authors

中国图像图形学报, Oct 2017

Bib HTML

@article{20170416,
  title = {正交投影非负矩阵的交替方向乘子分解方法},
  journal = {中国图像图形学报},
  volume = {22},
  number = {4},
  pages = {463-471},
  year = {2017},
  issn = {1006-8961},
  doi = {10.11834/jig.20170406},
  author = {周健 and 窦云峰 and 刘荣敏 and 王华彬 and 陶亮},
}

L1/2稀疏约束卷积非负矩阵分解的单通道语音增强方法

路成, 田猛, 周健, and 2 more authors

声学学报, Oct 2017

Bib HTML

@article{20170315,
  title = {L1/2稀疏约束卷积非负矩阵分解的单通道语音增强方法},
  journal = {声学学报},
  volume = {42},
  number = {3},
  pages = {377-384},
  year = {2017},
  issn = {0371-0025},
  doi = {10.15949/j.cnki.0371-0025.2017.03.016},
  author = {路成 and 田猛 and 周健 and 王华彬 and 陶亮},
}

采用稀疏和平滑双约束的增量正交映射非负矩阵分解目标跟踪

王华彬, 田猛, 周健, and 2 more authors

计算机辅助设计与图形学学报, Jan 2017

Bib HTML

@article{2017299,
  author = {王华彬 and 田猛 and 周健 and 施汉琴 and 陶亮},
  title = { 采用稀疏和平滑双约束的增量正交映射非负矩阵分解目标跟踪 },
  journal = {计算机辅助设计与图形学学报},
  year = {2017},
  volume = {29},
  number = {9},
  pages = {1658-1666},
  month = jan,
}

2016

联合模板先验概率和稀疏表示的目标跟踪

田猛, 路成, 周健, and 2 more authors

中国图象图形学报, Jan 2016

Bib HTML

@article{20162111,
  author = {田猛 and 路成 and 周健 and 施汉琴 and 陶亮},
  title = {联合模板先验概率和稀疏表示的目标跟踪},
  journal = {中国图象图形学报},
  year = {2016},
  volume = {21},
  number = {11},
  pages = {1455-1463},
  month = jan,
}

2015

仿人耳听觉的助听器双耳声源定位算法

梁瑞宇, 周健, 王青云, and 2 more authors

声学学报, Jan 2015

Bib HTML

@article{2015403,
  author = {梁瑞宇 and 周健 and 王青云 and 奚吉 and 赵力},
  title = {仿人耳听觉的助听器双耳声源定位算法 },
  journal = {声学学报},
  year = {2015},
  volume = {40},
  number = {3},
  pages = {446-454},
  month = jan,
}

非对称代价函数的稀疏卷积非负矩阵分解方法

张倩敏, 陶亮, 周健, and 1 more author

信号处理, Jan 2015

Bib HTML

@article{2015195102,
  author = {张倩敏 and 陶亮 and 周健 and 王华彬},
  title = {非对称代价函数的稀疏卷积非负矩阵分解方法},
  journal = {信号处理},
  year = {2015},
  number = {1},
  pages = {95-102},
  month = jan,
}

2014

Unsupervised learning of phonemes of whispered speech in a noisy environment based on convolutive non-negative matrix factorization

Jian Zhou, Ruiyu Liang, Li Zhao, and 2 more authors

Information Sciences, Jan 2014

Abs Bib HTML

This paper focuses on the development of an algorithm that can be optimized for a specific acoustic environment to improve the intelligibility of whispered speech. A new convolutive non-negative matrix factorization (NMF) algorithm is proposed to extract phoneme bases from noisy whispered speech with the noise bases from prior learning; these noise bases are obtained from training using the conventional non-negative matrix factorization. The divergence function with a sparseness constraint term is selected as the objective function in the developed algorithm to obtain multiplicative update rules of the phoneme base matrix and the corresponding weight matrix. The weights of the noise bases from prior learning are also updated in the phoneme learning stage. Listening experiments were conducted to assess the intelligibility performance of speech synthesized using the proposed algorithm. The experimental results indicate that the proposed algorithm is very effective for improving the intelligibility of whispers in various noise contexts, and it outperforms conventional algorithms.
@article{ZHOU2014115, title = {Unsupervised learning of phonemes of whispered speech in a noisy environment based on convolutive non-negative matrix factorization}, journal = {Information Sciences}, volume = {257}, pages = {115-126}, year = {2014}, issn = {0020-0255}, doi = {https://doi.org/10.1016/j.ins.2013.09.037}, author = {Zhou, Jian and Liang, Ruiyu and Zhao, Li and Tao, Liang and Zou, Cairong}, keywords = {Unsupervised phoneme base learning, Non-negative matrix factorization, Whisper intelligibility enhancement}, }

2012

Whisper Intelligibility Enhancement Using a Supervised Learning Approach

Jian Zhou, Liang Ruiyu, Li Zhao, and 1 more author

Circuits, Systems, and Signal Processing, Dec 2012

Bib HTML

@article{20121231,
  author = {Zhou, Jian and Ruiyu, Liang and Zhao, Li and Zou, Cairong},
  year = {2012},
  month = dec,
  title = {Whisper Intelligibility Enhancement Using a Supervised Learning Approach},
  volume = {31},
  journal = {Circuits, Systems, and Signal Processing},
  doi = {10.1007/s00034-012-9415-0},
}