语音增强(SE)
文章平均质量分 94
语音增强(SE)
凌逆战
保持真诚和善良,学会成熟,早睡早起,寻回热爱与运动,实现自我成就。关注我,我们就是朋友,互相进步呀
展开
-
语音增强论文翻译:2017_SEGAN: Speech Enhancement Generative Adversarial Network
论文地址:基于生成对抗网络的语音增强博客地址(转载请指明出处):https://www.cnblogs.com/LXP-Never/p/9986744.htmlSEGAN的例子摘要当前的语音增强技术是在频谱域或利用一些高级特征的基础上进行的。他们中的大多数人处理的噪音条件有限,并依赖于一阶统计特性。为了规避这些问题,深层网络正被越来越多地使用,因为它们能够从大量的训练数据集中学习到复杂...原创 2018-12-16 12:01:00 · 561 阅读 · 0 评论 -
SEGAN: Speech Enhancement Generative Adversarial Network
论文原文地址,目录摘要一、引言二、Generative Adversarial Networks三、Speech Enhancement GAN四、实验步骤4.1 数据集4.2 SEGAN步骤五、结果4.1 客观评价4.2 主观评价六、总结七、致谢八、参考文献摘要当前的语音增强技术是在频谱域或利用一些高级特征的基础上进行的。他们中...原创 2018-11-20 00:11:00 · 11993 阅读 · 22 评论 -
论文翻译:2021_FullSubNet: A Full-Band And Sub-Band Fusion Model For Real-Time Single-Channel Speech Enha...
论文地址:Fullsubnet:实时单通道语音增强的全频带和子频带融合模型代码地址:https://github.com/haoxiangsnr/FullSubNet引用格式:Hao X, Su X, Horaud R, et al. FullSubNet: A Full-Band and Sub-Band Fusion Model for Real-Time Single-Channel...原创 2021-11-10 11:59:00 · 985 阅读 · 0 评论 -
论文翻译:2021_语音增强模型压缩_Performance optimizations on deep noise suppression models...
我们研究了量级结构剪枝以加快深度噪声抑制(DNS)模型的推理时间。尽管深度学习方法在提高音频质量方面取得了显著的成功,但它们增加的复杂性阻碍了它们在实时应用中的部署。我们在基线上实现了7.25倍的推理加速,同时平滑了模型的性能退化。消融研究表明,我们提出的网络再参数化(即每层尺寸)是加速的主要驱动因素,而量级结构剪枝与直接训练较小尺寸的模型相比具有相当大的作用。我们报告推理速度,因为参数减少并不需要加速,并且我们使用精确的非侵入性客观语音质量度量来度量模型质量。......转载 2022-04-09 23:11:00 · 597 阅读 · 0 评论 -
论文翻译:2020_DTLN:Dual-Signal Transformation LSTM Network for Real-Time Noise Suppression
论文地址:双路信号变换LSTM网络的实时噪声抑制论文代码:https://github.com/breizhn/DTLN引用格式:Westhausen N L, Meyer B T. Dual-signal transformation LSTM network for real-time noise suppression[J]. arXiv preprint arXiv:2005.07...原创 2022-03-07 11:12:00 · 1190 阅读 · 0 评论 -
论文翻译:2018_CRN_A Convolutional Recurrent Neural Network for Real-Time Speech Enhancement
论文地址:用于实时语音增强的卷积递归神经网络代码地址:https://github.com/JupiterEthan/CRN-causal作者主页:https://jupiterethan.github.io/引用格式:Tan K, Wang D L. A Convolutional Recurrent Neural Network for Real-Time Speech Enhanc...原创 2021-12-08 18:31:00 · 853 阅读 · 0 评论 -
论文翻译:2020_Improving Perceptual Quality By Phone-Fortified Perceptual Loss For Speech Enhancement...
论文地址:通过语音增强的电话强化感知损失提高感知质量论文代码:https://github.com/aleXiehta/PhoneFortifiedPerceptualLoss引用格式:Hsieh T A, Yu C, Fu S W, et al. Improving Perceptual Quality by Phone-Fortified Perceptual Loss using W...原创 2021-12-09 08:11:00 · 546 阅读 · 0 评论 -
论文翻译:2020_FLGCNN: A novel fully convolutional neural network for end-to-end monaural speech enhancem...
论文地址:FLGCNN:一种新颖的全卷积神经网络,用于基于话语的目标函数的端到端单耳语音增强论文代码:https://github.com/LXP-Never/FLGCCRN(非官方复现)引用格式:Zhu Y, Xu X, Ye Z. FLGCNN: A novel fully convolutional neural network for end-to-end monaural spe...原创 2022-01-12 10:48:00 · 624 阅读 · 0 评论 -
论文翻译:2020_SEWUNet:Monaural Speech Enhancement Through Deep Wave-U-Net
论文地址:基于深度波形U-Net进行单耳语音增强论文代码:https://github.com/Hguimaraes/SEWUNet引用格式:Guimarães H R, Nagano H, Silva D W. Monaural speech enhancement through deep wave-U-net[J]. Expert Systems with Applications,...原创 2021-12-01 18:48:00 · 487 阅读 · 0 评论 -
麦克风阵列论文翻译:Speech Enhancement Based on the General Transfer Function GSC and Postfiltering
在语音增强应用中,麦克风阵列后置滤波可进一步减少波束形成器输出处的噪声成分。在麦克风阵列结构中,最近提出的通用传递函数广义旁瓣消除器(TF-GSC)在定向噪声场中显示出令人印象深刻的降噪能力,同时仍保持低语音失真。但是,在扩散噪声场中,可获得的降噪效果不明显。当噪声信号不稳定时,性能甚至会进一步下降。 在本文中,我们提出了三种后置滤波方法,以改善麦克风阵列的性能。 其中两个基于单通道语音增强器,并利用了最近提出的与波束形成器输出串联的算法。 第三个是多通道语音增强器,它利用TF-GSC结构中构建的纯噪声组件转载 2020-02-25 09:16:00 · 1482 阅读 · 3 评论 -
论文翻译:2021_PercepNet:A Perceptually Motivated Approach for Low-complexity, Real-time Enhancement of F...
论文地址:一种低复杂度实时增强全频带语音的感知激励方法论文代码:https://github.com/search?q=PercepNet引用格式:Valin J M, Isik U, Phansalkar N, et al.A Perceptually Motivated Approach for Low-complexity, Real-time Enhancement of Ful...原创 2021-12-12 17:00:00 · 660 阅读 · 0 评论 -
论文翻译:2021_Low-Delay Speech Enhancement Using Perceptually Motivated Target and Loss
论文地址:使用感知动机目标和损失的低延迟语音增强引用格式:Zhang X, Ren X, Zheng X, et al. Low-Delay Speech Enhancement Using Perceptually Motivated Target and Loss[J]. Proc. Interspeech 2021, 2021: 2826-2830.摘要 基于深度神经网络的语音...原创 2021-12-13 11:18:00 · 550 阅读 · 2 评论 -
论文翻译:2022_PACDNN: A phase-aware composite deep neural network for speech enhancement
论文地址:PACDNN:一种用于语音增强的相位感知复合深度神经网络相似代码:https://github.com/phpstorm1/SE-FCN引用格式:Hasannezhad M,Yu H,Zhu W P,et al. PACDNN: A phase-aware composite deep neural network for speech enhancement[J]. Speec...原创 2022-02-15 15:39:00 · 1170 阅读 · 0 评论 -
论文翻译:2020_Demucs:Real Time Speech Enhancement in the Waveform Domain
论文地址:在波形域的实时语音增强论文作者:facebook AI 研究所论文代码:https://github.com/facebookresearch/denoiser摘要 我们提出了一个基于原始波形的因果语音增强模型,该模型在笔记本电脑CPU上实时运行。所提出的模型是基于一个带有跳跃连接的编码器-解码器架构。利用多个损耗函数,在时域和频域上都得到了优化。实验结果表明,该方法能够去...原创 2021-11-17 19:50:00 · 1458 阅读 · 0 评论 -
论文翻译:2020_TinyLSTMs: Efficient Neural Speech Enhancement for Hearing Aids
论文地址:TinyLSTMs:助听器的高效神经语音增强音频地址:https://github.com/Bose/efficient-neural-speech-enhancement引用格式:Fedorov I,Stamenovic M,Jensen C,et al. TinyLSTMs:Efficient neural speech enhancement for hearing aid...原创 2022-04-18 12:00:00 · 607 阅读 · 1 评论 -
论文翻译:2021_语音增强模型压缩_Towards model compression for deep learning based speech enhancement...
论文地址:面向基于深度学习的语音增强模型压缩论文代码:没开源,鼓励大家去向作者要呀,作者是中国人,在语音增强领域 深耕多年引用格式:Tan K, Wang D L. Towards model compression for deep learning based speech enhancement[J]. IEEE/ACM transactions on audio, speech, ...原创 2022-04-08 10:58:00 · 879 阅读 · 0 评论 -
论文翻译:2019_TCNN: Temporal convolutional neural network for real-time speech enhancement in the time d...
论文地址:TCNN:时域卷积神经网络用于实时语音增强论文代码:https://github.com/LXP-Never/TCNN(非官方复现)引用格式:Pandey A, Wang D L. TCNN: Temporal convolutional neural network for real-time speech enhancement in the time domain[C]//...原创 2022-01-18 17:42:00 · 702 阅读 · 0 评论 -
论文翻译:2020_Densely connected neural network with dilated convolutions for real-time speech enhancemen...
论文名称:扩展卷积密集连接神经网络用于时域实时语音增强论文代码:https://github.com/ashutosh620/DDAEC引用:Pandey A, Wang D L. Densely connected neural network with dilated convolutions for real-time speech enhancement in the time d...原创 2021-11-26 12:05:00 · 731 阅读 · 0 评论 -
论文翻译:2020_NSNet:Weighted speech distortion losses for neural-network-based real-time speech enhancem...
论文地址:基于神经网络的实时语音增强的加权语音失真损失论文代码:https://github.com/GuillaumeVW/NSNet引用:Xia Y, Braun S, Reddy C K A, et al. Weighted speech distortion losses for neural-network-based real-time speech enhancement[C...原创 2021-12-06 15:25:00 · 604 阅读 · 0 评论 -
论文翻译:2021_MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement
论文地址:MetricGAN+:用于语音增强的 MetricGAN 的改进版本论文代码:https://github.com/JasonSWFu/MetricGAN引用格式:Fu S W, Yu C, Hsieh T A, et al. MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement[J]. arXiv...原创 2021-12-21 17:02:00 · 2563 阅读 · 0 评论 -
论文翻译:2020_DARCN_A Recursive Network with Dynamic Attention for Monaural Speech Enhancement
论文地址:基于动态注意的递归网络单耳语音增强论文代码:https://github.com/Andong-Li-speech/DARCN引用格式:Li, A., Zheng, C., Fan, C., Peng, R., Li, X. (2020) A Recursive Network with Dynamic Attention for Monaural Speech Enhancem...原创 2021-12-01 16:01:00 · 657 阅读 · 0 评论 -
论文翻译:2020_RNNoise:A Hybrid DSP/Deep Learning Approach to Real-Time Full-Band Speech Enhancement...
网上已经有很多人翻译了,但我做这工作只是想让自己印象更深刻文章方向:语音增强论文地址:基于DSP/深度学习的实时全频带语音增强方法博客地址:https://www.cnblogs.com/LXP-Never/p/15144882.html论文代码:https://github.com/xiph/rnnoise主页:https://jmvalin.ca/demo/rnnoise/摘要...原创 2021-08-16 20:07:00 · 1105 阅读 · 0 评论 -
论文翻译:2020_WaveCRN: An efficient convolutional recurrent neural network for end-to-end speech enhance...
论文地址:用于端到端语音增强的卷积递归神经网络论文代码:https://github.com/aleXiehta/WaveCRN引用格式:Hsieh T A, Wang H M, Lu X, et al. WaveCRN: An efficient convolutional recurrent neural network for end-to-end speech enhancemen...原创 2021-11-23 17:47:00 · 493 阅读 · 0 评论 -
论文翻译:2020_DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement...
论文地址:DCCRN:用于相位感知语音增强的深度复杂卷积循环网络论文代码:https://paperswithcode.com/paper/dccrn-deep-complex-convolution-recurrent-1引用:Hu Y,Liu Y,Lv S,et al. DCCRN: Deep complex convolution recurrent network for phas...原创 2022-03-09 15:23:00 · 821 阅读 · 0 评论 -
论文翻译:2021_DeepFilterNet: A Low Complexity Speech Enhancement Framework for Full-Band Audio based on ...
论文地址:DeepFilterNet:基于深度滤波器的全频带音频低复杂度语音增强框架论文代码:https://github.com/Rikorose/DeepFilterNet引用:Schröter H, Rosenkranz T, Maier A. DeepFilterNet: A Low Complexity Speech Enhancement Framework for Full-...原创 2022-01-20 21:21:00 · 951 阅读 · 0 评论