周琳

办公室：九龙湖校区电子信息大楼A318

Email：Linzhou@seu.edu.cn

学习经历：

2013/7-2013/8   美国里海大学访学

2000/9-2005/7   779cn太阳集团硕博连读

1996/9-2000/6   779cn太阳集团本科

工作经历:

2021/7-   至今779cn太阳集团副教授、博导

2009/4-2021/7     779cn太阳集团副教授、硕导

2005/7-2009/4     779cn太阳集团讲师

教授课程：

人工智能与深度学习—本科三年级（春季），必修课，48学时

语音信号处理—本科三年级（秋季），限选课，48学时

MATLAB实践—本科三年级（暑期），必修课，32学时

研究方向：

1、深度学习、模式识别与机器学习

2、语音信号处理，包括语音识别、语音增强、语音编码

3、声学信号处理，包括声源定位、声源分离

主持省重点研发课题、国家重点研发计划子课题以及国家自然科学基金面上项目等项目。以第一作者和通讯作者发表论文50余篇，SCI收录30余篇，授权和公开发明专利20余项。

获奖情况：

1、2008年779cn太阳集团青年授课竞赛

2、2014年779cn太阳集团奖教金

论文著作：

[1]Lin ZHOU, Tongjia YAN, Mingyang LI, Ao LIs. Monaural Speech   Enhancement with Attention Augmented Dual-Path CRN and Short Time Discrete   Cosine Transform. IEICE Transactions on Fundamentals of Electronics,   Communications and Computer Sciences. Vol.E108-A, No.11, pp 1444-1451., Nov.   2025.

[2]Lin ZHOU, Yanxiang CAO, Qirui WANG, Yunling CHENG, Chenghao   ZHUANG, and Yuxi DENG. Monaural Speech Enhancement with Full-Convolution   Attention Module and Post-Processing Strategy. IEICE Transactions on   Fundamentals of Electronics, Communications and Computer Sciences. 2025:   Vol.E108-A, No.8. pp.996-1004

[3]Lin Zhou, Qirui Wang, Haochen Bao, Yonghong Peng, and Xu Cheng.   Distortion Discriminator Helps Improve Speech Restoration. IEEE Signal   Processing Letters. 2025.

[4]Cheng,   Jiaming; Liang, Ruiyu; Zhou, Lin; Zhao, Li; Huang, Chengwei; Schuller,   Bjorn W. Residual Fusion Probabilistic Knowledge Distillation for Speech   Enhancement. IEEE/ACM Transactions on Audio Speech and Language Processing,   2024, 32, pp:2680-2691.

[5]Cong PANG,   Ye NI, Jia Ming CHENG, Lin ZHOU, and Li ZHAO. BP-CRN: A Lightweight   Two-Stage Convolutional Recurrent Network for Multi-channel Speech   Enhancement. IEICE Transactions on Information and Systems. 2024.

[6]Cong Pang,   Feifei Xiong, Ye Ni, Lin Zhou, Jinwei Feng. The SEUEE System for the   CHiME-8 MMCSG Challenge. Pang C, Xiong F, Ni Y, et al. The SEUEE System for   the CHiME-8 MMCSG Challenge. 2024.

[7]Wang, Qirui,   Zhou, Lin; Cao, Yanxiang; Zhuang, Chenghao; Cheng, Yunling; Deng, Yux.   I. A Low Computation Cost Model for Real-Time Speech Enhancement. 2024 IEEE   13th International Conference on Communications, Circuits, and Systems, 2024,   pp:267-271.

[8]Zhuang,   Chenghao, Zhou, Lin; Cao, Yanxiang; Wang, Qirui; Cheng, Yunling.   DCE-CDPPTnet: Dense Connected Encoder Cross Dual-Path Parrel Transformer   Network for Multi-Channel Speech Separation. 2024 IEEE 13th International   Conference on Communications, Circuits, and Systems, 2024, pp:303-308.

[9]Cheng   Yunling, Zhou Lin, Wang qirui, Cao Yanxiang, Zhang Chenghao. A speech   enhancement method based on dual-path phase-aware GAN networks. 2024 IEEE   13th International Conference on Communications, Circuits, and Systems, 2024,   pp:315-320, 2024.

[10]Wang, Jie;   Ju, Mengjie; Zhou, Lin. Self-Fitting Hearing Aid Algorithm Based on   Improved Interactive Genetic Algorithm.    Mechanisms and Machine Science, Advances in Automation, Mechanical and   Design Engineering – 2023,138:277-289.

[11]Wang, Jie,   Ju, Mengjie; Zhou, Lin. Research on Intelligent Fitting Algorithm for   OTC Hearing Aids. ACM International Conference Proceeding Series, p 220-224,   2023.

[12]Lin Zhou, Xijin   Chen, Chaoyan Wu, Qiuyue Zhong, Xu Cheng, Yibin Tang.Speech Enhancement via Mask-Mapping Based Residual   Dense Network. Computers, Materials & Continua .2023, 74(1), 1259-1277.

[13]Zhou L, Zhong Q,   Wang T, et al. Speech Enhancement via Residual Dense Generative Adversarial   Network. Computer Systems Science and Engineering, 2021, 38(3):279-289.

[14]Zhou L, Xu Y, Wang   TY, Feng K, et al. Microphone Array Speech Separation Algorithm based on   TC-ResNet. CMC-Computers, Materials & Continua.2021, 69(2):2705-2716.

[15]Zhou L, Feng K,   Wang TY, Xu Y, et al. Binaural Speech Separation Algorithm Based on Deep   Clustering. Intelligent Automation & Soft Computing. 2021, 30(2):527-537.

[16]Wu C, Zhou L, Chen X, et al. Microphone   Array Speech Separation Algorithm based on DNN[C]//2021 Asia-Pacific Signal   and Information Processing Association Annual Summit and Conference. IEEE,   2021: 1305-1310.

[17]Cheng,   X;Zhang, YF;Zhou, L;Lu, GJ. Online   dual dictionary learning for visual object tracking. Journal of Ambient   Intelligence and Humanized Computing.2021,   10881–10896.

[18]Tang,   YB;Sun, J;Jiang, AM;Chen, Y;Zhou, L.   Adaptive Graph Filtering with Intra-Patch Pixel Smoothing for Image   Denoising. Circuits Systems and Signal Processing. 2021, 40(11):5381-5400

[19]Lin Zhou, Siyuan Lu,   Qiuyue Zhong, Ying Chen, Yibin Tang, Yan Zhou. Binaural Speech Separation   Algorithm Based on Long and Short Time Memory Networks. CMC, vol.63, no.3,   pp.1373-1386, 2020.

[20]Cheng, X; Zhang,   YF; Zhou, L; Zheng, YH. Visual   Tracking via Auto-Encoder Pair Correlation Filter. IEEE Transactions on   Industrial Electronics. 2020, 67(4):3288-3297.

[21]Lin Zhou, Kangyu Ma,   Lijie Wang, Ying Chen and Yibin Tang. Binaural Sound Source Localization   Based on Convolutional Neural Network. Computers, Materials & Continua.   2019，60(2):545-557.

科研项目：

项目名称

项目类别

项目时间

项目金额

承担情况

平面阵列麦克风开发

横向课题

2026.2-2028.2

30万

主持

远场环境下的语音检测和增强关键技术研发

江苏省重点研发课题

2022.6-2026.6

300万

主持

技术类

国家级课题

2022.12-2025.12

200万

参与

助听器免验配关键技术研究/助听器交互式智能验配方法研究及实现

国家重点研发子课题

2020.7-2023.6

20万

主持

科技类

横向课题

2021.12-2023.12

49

参与

安全监控网络系统

企业合作

2020.12-2023.12

78万

主持

基于空间听觉感知的双耳语音分离和识别关键问题研究

国家自然科学基金

2016.01-2019.12

68万

主持

双耳交互计算模型与空间听觉研究

国家自然科学基金

2013.01-2015.12

25万

主持

****

总装预研项目

2012.01-2015.12

60万

主持

专利：

专利号

专利名称

专利类型

202211155820.9

一种基于FirePS卷积神经网络的语音增强方法

授权专利

202111192259.7

一种基于掩膜映射和混合空洞卷积网络的语音增强方法

授权专利

202011285987.8

基于对数幅度谱和耳间相位差的深度聚类语音分离方法

授权专利

202011251485.3

基于TC-ResNet网络的麦克风阵列语音分离方法

授权专利

202010872003.X

基于Soft-argmax回归器的双耳声源定位方法

授权专利

202010835547.9.

一种基于密集卷积网络的双耳声源定位方法

授权专利

201910896903.5

一种基于二维语谱图和条件生成对抗网络的语音增强方法

授权专利

201910930176.X

一种基于长短时记忆网络LSTM的双耳语音分离方法

授权专利

202211515165.3

基于简化注意力编解码网络的滤波求和多通道语音分离方法

受理专利

202211155820.9

一种基于FirePS卷积神经网络的语音增强方法

受理专利

202211294392.8

一种基于卷积自注意力编码结构的语音增强方法

受理专利