利用生成噪声提高语音增强方法的泛化能力

doi:10.3969/j.issn.0372-2112.2019.04.004

电子学报 ›› 2019, Vol. 47 ›› Issue (4): 791-797.DOI: 10.3969/j.issn.0372-2112.2019.04.004

利用生成噪声提高语音增强方法的泛化能力

袁文浩, 娄迎曦, 梁春燕, 夏斌

山东理工大学计算机科学与技术学院, 山东淄博 255000

收稿日期:2018-02-27 修回日期:2018-07-24 出版日期:2019-04-25
- 通讯作者:
- 袁文浩
- 作者简介:
- 娄迎曦 女,1996年出生,山东聊城人.2018年毕业于青岛理工大学琴岛学院,现为山东理工大学计算机科学与技术学院硕士研究生.主要研究方向为语音信号处理、语音增强.E-mail:1804224373@qq.com
- 基金资助:
- 国家自然科学基金 (No.61701286，No.11704229）; 山东省自然科学基金 (No.ZR2015FL003，No.ZR2017MF047，No.ZR2017LA011）

Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise

YUAN Wen-hao, LOU Ying-xi, LIANG Chun-yan, XIA Bin

College of Computer Science and Technology, Shandong University of Technology, Zibo, Shandong 255000, China

Received:2018-02-27 Revised:2018-07-24 Online:2019-04-25 Published:2019-04-25
- Supported by:
- National Natural Science Foundation of China (No.61701286, No.11704229); Natural Science Foundation of Shandong Province, China (No.ZR2015FL003, No.ZR2017MF047, No.ZR2017LA011)

摘要/Abstract

摘要： 如何提高对未知噪声类型的泛化能力是有监督语音增强方法中亟待解决的重要问题，通过对大量不同类型噪声进行建模，深度神经网络成为了解决该问题的有效手段.为了进一步提高基于深度神经网络的语音增强方法的泛化能力，本文基于生成式对抗网络（Generative Adversarial Networks，GAN）设计了能够由真实噪声数据生成新的噪声类型的NoiseGAN；通过在训练集中增加生成噪声类型，提高训练集噪声类型的多样性，从而达到提高语音增强模型泛化能力的目的.不同结构的网络下的语音增强实验结果表明，本文提出的NoiseGAN能够生成新的噪声类型，具备提高训练集噪声类型多样性的能力，有效提高了语音增强模型在未知噪声类型下的泛化能力.

关键词: 语音增强, 生成式对抗网络, 泛化能力, 深度神经网络

Abstract: How to improve the generalization ability to unknown noise types is an important problem to be solved urgently in supervised speech enhancement approaches.By modeling a large number of types of noise,the deep neural network(DNN)becomes an effective way to solve this problem.In order to further improve the generalization ability of speech enhancement approaches based on DNN,this paper designs NoiseGAN based on Generative Adversarial Networks (GAN) to generate new noise types from real noise data.By adding generated noise to training set,the diversity of noise types in training set is increased,and thereby the generalization ability of speech enhancement model is improved.The results of speech enhancement experiments under different structures of networks show that the proposed NoiseGAN can generate new noise types,increase the diversity of noise types in training set,and effectively improve the generalization ability of speech enhancement models under unknown noise types.

Key words: speech enhancement, generative adversarial networks, generalization ability, deep neural network

中图分类号:

TN912.3

袁文浩, 娄迎曦, 梁春燕, 等. 利用生成噪声提高语音增强方法的泛化能力[J]. 电子学报, 2019, 47(4): 791-797.

YUAN Wen-hao, LOU Ying-xi, LIANG Chun-yan, et al. Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise[J]. Acta Electronica Sinica, 2019, 47(4): 791-797.

参考文献

[1] LOIZOU P C.Speech Enhancement:Theory and Practice[M].UK:CRC Press,2013.
[2] 孟宪波,鲍长春.基于最小控制GARCH模型的噪声估计算法[J].电子学报,2016,44(3):747-752. MENG Xian-bo,BAO Chang-chun.Noise estimate algorithm based on minima controlled GARCH model[J].Acta Electronica Sinica,2016,44(3):747-752.(in Chinese)
[3] 何玉文,鲍长春,夏丙寅,等.基于AR-HMM在线能量调整的语音增强方法[J].电子学报,2014,42(10):1991-1997. HE Yu-wen,BAO Chang-chun,XIA Bing-yin,et al.Online energy adjustment using AR-HMM for speech enhancement[J].Acta Electronica Sinica,2014,42(10):1991-1997.(in Chinese)
[4] MOHAMMADIHA N,SMARAGDIS P,LEIJON A.Supervised and unsupervised speech enhancement using nonnegative matrix factorization[J].IEEE Transactions on Audio,Speech,and Language Processing,2013,21(10):2140-2151.
[5] 刘文举,聂帅,梁山,等.基于深度学习语音分离技术的研究现状与进展[J].自动化学报,2016,42(6):819-833. LIU Wen-Ju,NIE Shuai,LIANG Shan,et al.Deep learning based speech separation technology and its developments[J].Acta Automatica Sinica,2016,42(6):819-833.(in Chinese)
[6] XU Y,DU J,DAI L R,et al.An experimental study on speech enhancement based on deep neural networks[J].IEEE Signal Processing Letters,2014,21(1):65-68.
[7] XU Y,DU J,DAI L R,et al.A regression approach to speech enhancement based on deep neural networks[J].IEEE/ACM Transactions on Audio,Speech,and Language Processing,2015,23(1):7-19.
[8] XU Y,DU J,HUANG Z,et al.Multi-objective learning and mask-based post-processing for deep neural network based speech enhancement[A].Proceedings of Sixteenth Annual Conference of the International Speech Communication Association[C].Dresden:ISCA,2015.1508-1512.
[9] WANG Y,CHEN J,WANG D L.Deep Neural Network Based Supervised Speech Segregation Generalizes to Novel Noises Through Large-Scale Training[R].Ohio State University Columbus,2015.
[10] CHEN J,WANG Y,YOHO S E,et al.Large-scale training to increase speech intelligibility for hearing-impaired listeners in novel noises[J].The Journal of the Acoustical Society of America,2016,139(5):2604-2612.
[11] CHEN J,WANG Y,WANG D L.Noise perturbation for supervised speech separation[J].Speech Communication,2016,78:1-10.
[12] GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial nets[A].Proceedings of Advances in Neural Information Processing Systems[C].US:NIPS,2014.2672-2680.
[13] RADFORD A,METZ L,CHINTALA S.Unsupervised representation learning with deep convolutional generative adversarial networks[A].International Conference on Learning Representations[C].US:ICLR,arXiv:1511.06434v2.
[14] HU G.100 Nonspeech Environmental Sounds,2004[OL].http://web.cse.ohio-state.edu/pnl/corpus/HuNonspeech/HuCorpus.html,2004.
[15] GAROFOLO J S,LAMEL L F,FISHER W M,et al.TIMIT Acoustic-Phonetic Continuous Speech Corpus[CD].Philadelphia:Linguistic Data Consortium,1993.
[16] VARGA A,STEENEKEN H J M.Assessment for automatic speech recognition:Ⅱ.NOISEX-92:A database and an experiment to study the effect of additive noise on speech recognition systems[J].Speech Communication,1993,12(3):247-251.
[17] YU D,EVERSOLE A,SELTZER M,et al.An Introduction to Computational Networks and the Computational Network Toolkit[R].Tech Rep MSR,Microsoft Research,2014.
[18] RIX A W,BEERENDS J G,HOLLIER M P,et al.Perceptual evaluation of speech quality (PESQ)-a new method for speech quality assessment of telephone networks and codecs[A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing[C].US:IEEE,2001.749-752.
[19] TAAL C H,HENDRIKS R C,HEUSDENS R,et al.An algorithm for intelligibility prediction of time-frequency weighted noisy speech[J].IEEE Transactions on Audio,Speech,and Language Processing,2011,19(7):2125-2136.

利用生成噪声提高语音增强方法的泛化能力

Improving Generalization Ability of Speech Enhancement Approaches Using Generated Noise

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	王硕, 王坚, 王亚男, 宋亚飞. 一种基于特征融合的恶意代码快速检测方法[J]. 电子学报, 2023, 51(1): 57-66.
[2]	周静, 鲍长春, 张旭. 基于聚焦信号子空间估计导向矢量的干扰声源抑制方法[J]. 电子学报, 2023, 51(1): 76-85.
[3]	佘维, 霍丽娟, 刘炜, 张志鸿, 宋轩, 田钊. 一种可隐藏敏感文档和发送者身份的区块链隐蔽通信模型[J]. 电子学报, 2022, 50(4): 1002-1013.
[4]	周奕涛, 张斌, 刘自豪. 基于多模态深度神经网络的应用层DDoS攻击检测模型[J]. 电子学报, 2022, 50(2): 508-512.
[5]	陈星宇, 叶锋, 黄添强, 翁彬, 陈家祯, 林文忠. 融合小型深度生成模型的显著性检测[J]. 电子学报, 2021, 49(4): 768-774.
[6]	韩冲, 王俊丽, 吴雨茜, 张超波. 基于神经进化的深度学习模型研究综述[J]. 电子学报, 2021, 49(2): 372-379.
[7]	黎毅达, 高发荣, 姚婷, 蔡利杰. 基于Fisher判别和GKF-RELM算法的多特征步态模式识别[J]. 电子学报, 2021, 49(10): 1993-2001.
[8]	崔子豪, 鲍长春. 基于广义合成分析和深度神经网络的自回归系数估计方法[J]. 电子学报, 2021, 49(1): 29-39.
[9]	欧世峰, 赵艳磊, 宋鹏, 高颖. 基于概率耦合的双直接判决先验信噪比估计算法[J]. 电子学报, 2020, 48(8): 1605-1614.
[10]	袁文浩, 胡少东, 时云龙, 李钊, 梁春燕. 一种用于语音增强的卷积门控循环网络[J]. 电子学报, 2020, 48(7): 1276-1283.
[11]	张嘉纹, 党小宇, 杨凌辉, 齐茹梦, 王萌. 海面短波地波通信中基于DNN神经网络的单样本极化滤波器预测研究[J]. 电子学报, 2020, 48(11): 2250-2257.
[12]	袁子晗, 蒋明峰, 李杨, 支明豪, 朱志军. 基于DAWGAN-GP的磁共振图像重构方法研究[J]. 电子学报, 2020, 48(10): 1883-1890.
[13]	沈卉卉, 刘国武, 付丽华, 刘智慧, 李宏伟. 一种基于修正动量的RBM算法[J]. 电子学报, 2019, 47(9): 1957-1964.
[14]	徐少平, 李崇禧, 林官喜, 唐祎玲, 胡凌燕. 基于主成分分析与深度神经网络的快速噪声水平估计算法[J]. 电子学报, 2019, 47(2): 274-281.
[15]	陈楠, 鲍长春. 基于双耳线索编码原理的语音增强方法[J]. 电子学报, 2019, 47(1): 227-233.