基于环状生成对抗网络的深度语音去噪方法
|
摘要
针对基于深度学习的语音信号去噪方法存在难于收敛、性能不足的问题,本文提出了基于环状生成对抗网络的深度语音信号去噪方法,设计了新型的环状生成对抗语义去噪网络。通过40余种不同噪声语音集的试验,结果表明所提方法在5种衡量标准下都明显改善了去噪性能。
Traditional deep learning based audio denoising methods are difficult to convergence and their performances are insufficient to practical applications. This paper proposes a new audio denoising algorithms by CycleGAN,and design a new audio denoising network. By verifying the proposed method on 40 different types of audio noises,the experimental results demonstrate that the proposed method outperforms the existing methods obviously on five evaluation metrics.
Traditional deep learning based audio denoising methods are difficult to convergence and their performances are insufficient to practical applications. This paper proposes a new audio denoising algorithms by CycleGAN,and design a new audio denoising network. By verifying the proposed method on 40 different types of audio noises,the experimental results demonstrate that the proposed method outperforms the existing methods obviously on five evaluation metrics.
引文
[1]郑尊凯,文畅,谢凯,等.ELMD联合粒子群优化小波阈值的语音去噪研究[J].长江大学学报:自科版,2018,15(1):33-38.
[2]李斌.数字助听器中语音增强算法的研究[D].南京:南京邮电大学,2017.
[3]王潇涵.语音信号去噪研究[J].数字技术与应用,2017(6):76-77,79.
[4]何珏杉.基于多通道语音增强和去噪的优化研究[D].南京:南京信息工程大学,2017.
[5] Kaizhi Qian,Yang Zhang,Shiyu Chang,et al.Speech enhancement using bayesian wavenet[C]//Interspeech,2017:2013-2017.
[6] Ravi Kumar Kandagatla,Subbaiah P V. Speech enhancement using MMSE estimation of amplitude and complexspeech spectral coecients under phase-uncertainty[C]//Speech Communication,2017:96-100.
[7]甘振业,陈浩,杨鸿武.结合EEMD与K-SVD字典训练的语音增强算法[J].清华大学学报:自然科学版,2017,57(3):286-292.
[8] Liu D,Smaragdis P,Kim M. Experiments on deep learning for speech denoising. Proceedings of the Annual Conference of the International Speech Communication Association[C]//Interspeech,2014:2685-2689.
[9] XU Yong,DU Jun,DAI Li-rong,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):7-19.
[10]Santiago Pascual,Antonio Bonafonte,Joan Serra.SEGAN:Speech Enhancement Generative Adversarial Network[C]//Interspeech,2017:3642-3646.
[11]OU Shifeng,SONG Peng,GAO Ying. Laplacian speech model and soft decision based MMSE estimator for noise power spectral density in speech enhancement[J]. Chinese Journal of Electronics,2018,27(6):1214-1220.
[12]Zhang Y,Shi L,Zhou Y. An improved perceptual KLT approach for speech enhancement[C]//International Conference on Computer Science&Network technology,2016:1419-1423.
[13]Dimitri Palaz,Mathew Magimai Doss,Ronan Collobert. Convolutional neural networks-based continuous speech recognition using raw speech signal[C]//ICASSP,2015:4295-4299.
[14]罗元,谭琴,张毅.基于Teager能量算子的改进阈值函数的去噪算法研究[J].计算机应用研究,2019(5):1-3.
[15]ZHU Jun-Yan,Taesung Park,Phillip Isola,et al.Unpaired image-to-image translation using cycleconsistent adversarial networks[C]//2017 IEEE International Conference on Computer Vision,2017:2223-2232.
[16]Shanthini Pandiaraj,Shankar Kumar K R. Speaker identification using discrete wavelet transform[J].Journal of Computer Science,2015(11):53-56.
[17]Noureddine Aloui, Souha Bousselmi, Adnane Cherif. Optimized speech compression algorithm based on wavelets techniques and its real time implementation on DSP[J]. International Journal of Information Technology and Computer Science,2015(3):33-41.
[18]胡杰,乔建华.基于NPEC-OMP的语音增强算法[J].太原科技大学学报,2017,38(5):342-348.
[19]陆振宇,何珏杉,赵为汉.关于多通道语音去噪的识别优化研究[J].计算机仿真,2016,33(6):315-320.
[20]钟斌,黄玉水,周明建.基于以太网的串行数字语音通信系统[J].现代电子技术,2019(3):28-30.
[21]邱鹏瑞,袁希平,甘淑,等.基于Android平台与QR码生成、识别技术的高校仪器设备管理系统研究[J].自动化与仪器仪表,2018(4):67-70.
[22]张莎莎.噪声背景下单稳系统的信号检测和恢复[J].纺织高校基础科学学报,2018(2):246-252.
[23]赵建平,李刚.基于GA-Elman神经网络的短期风电功率预测[J].陕西电力,2017(3):23-26.
[24]庄宿国,赵伟刚,杨霞辉,等.基于AutoCAD平台的涨圈密封参数化软件开发[J].火箭推进,2017(3):47-52.
[2]李斌.数字助听器中语音增强算法的研究[D].南京:南京邮电大学,2017.
[3]王潇涵.语音信号去噪研究[J].数字技术与应用,2017(6):76-77,79.
[4]何珏杉.基于多通道语音增强和去噪的优化研究[D].南京:南京信息工程大学,2017.
[5] Kaizhi Qian,Yang Zhang,Shiyu Chang,et al.Speech enhancement using bayesian wavenet[C]//Interspeech,2017:2013-2017.
[6] Ravi Kumar Kandagatla,Subbaiah P V. Speech enhancement using MMSE estimation of amplitude and complexspeech spectral coecients under phase-uncertainty[C]//Speech Communication,2017:96-100.
[7]甘振业,陈浩,杨鸿武.结合EEMD与K-SVD字典训练的语音增强算法[J].清华大学学报:自然科学版,2017,57(3):286-292.
[8] Liu D,Smaragdis P,Kim M. Experiments on deep learning for speech denoising. Proceedings of the Annual Conference of the International Speech Communication Association[C]//Interspeech,2014:2685-2689.
[9] XU Yong,DU Jun,DAI Li-rong,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):7-19.
[10]Santiago Pascual,Antonio Bonafonte,Joan Serra.SEGAN:Speech Enhancement Generative Adversarial Network[C]//Interspeech,2017:3642-3646.
[11]OU Shifeng,SONG Peng,GAO Ying. Laplacian speech model and soft decision based MMSE estimator for noise power spectral density in speech enhancement[J]. Chinese Journal of Electronics,2018,27(6):1214-1220.
[12]Zhang Y,Shi L,Zhou Y. An improved perceptual KLT approach for speech enhancement[C]//International Conference on Computer Science&Network technology,2016:1419-1423.
[13]Dimitri Palaz,Mathew Magimai Doss,Ronan Collobert. Convolutional neural networks-based continuous speech recognition using raw speech signal[C]//ICASSP,2015:4295-4299.
[14]罗元,谭琴,张毅.基于Teager能量算子的改进阈值函数的去噪算法研究[J].计算机应用研究,2019(5):1-3.
[15]ZHU Jun-Yan,Taesung Park,Phillip Isola,et al.Unpaired image-to-image translation using cycleconsistent adversarial networks[C]//2017 IEEE International Conference on Computer Vision,2017:2223-2232.
[16]Shanthini Pandiaraj,Shankar Kumar K R. Speaker identification using discrete wavelet transform[J].Journal of Computer Science,2015(11):53-56.
[17]Noureddine Aloui, Souha Bousselmi, Adnane Cherif. Optimized speech compression algorithm based on wavelets techniques and its real time implementation on DSP[J]. International Journal of Information Technology and Computer Science,2015(3):33-41.
[18]胡杰,乔建华.基于NPEC-OMP的语音增强算法[J].太原科技大学学报,2017,38(5):342-348.
[19]陆振宇,何珏杉,赵为汉.关于多通道语音去噪的识别优化研究[J].计算机仿真,2016,33(6):315-320.
[20]钟斌,黄玉水,周明建.基于以太网的串行数字语音通信系统[J].现代电子技术,2019(3):28-30.
[21]邱鹏瑞,袁希平,甘淑,等.基于Android平台与QR码生成、识别技术的高校仪器设备管理系统研究[J].自动化与仪器仪表,2018(4):67-70.
[22]张莎莎.噪声背景下单稳系统的信号检测和恢复[J].纺织高校基础科学学报,2018(2):246-252.
[23]赵建平,李刚.基于GA-Elman神经网络的短期风电功率预测[J].陕西电力,2017(3):23-26.
[24]庄宿国,赵伟刚,杨霞辉,等.基于AutoCAD平台的涨圈密封参数化软件开发[J].火箭推进,2017(3):47-52.