arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2204.03042cs.SDcs.AIeess.AS
收藏

FFC-SE:用于语音增强的快速傅立叶卷积

FFC-SE: Fast Fourier Convolution for Speech Enhancement

Samsung AI Center(三星人工智能中心);Higher School of Economics(高等经济学院);Skolkovo Institute of Science and Technology(斯科尔科沃科学技术研究院);Artificial Intelligence Research Institute(人工智能研究院)

Ivan Shchekotov, Pavel Andreev, Oleg Ivanov, Aibek Alanov, Dmitry Vetrov

摘要:

快速傅立叶卷积(FFC)是近年来提出的一种新的神经网络算子,在计算机视觉领域有着广阔的应用前景。FFC算子允许在神经网络的早期层内采用大的感受野操作。它被证明是特别有助于修复的周期性结构,这是常见的音频处理。在这项工作中,我们设计的神经网络架构,适应FFC语音增强。我们假设一个大的感受野允许这些网络产生比普通卷积模型更多的相干相位,并通过实验验证了这一假设。我们发现,基于快速傅立叶卷积的神经网络优于类似的卷积模型,并显示出更好或与其他语音增强基线相当的结果。

英文摘要:

Fast Fourier convolution (FFC) is the recently proposed neural operator showing promising performance in several computer vision problems. The FFC operator allows employing large receptive field operations within early layers of the neural network. It was shown to be especially helpful for inpainting of periodic structures which are common in audio processing. In this work, we design neural network architectures which adapt FFC for speech enhancement. We hypothesize that a large receptive field allows these networks to produce more coherent phases than vanilla convolutional models, and validate this hypothesis experimentally. We found that neural networks based on Fast Fourier convolution outperform analogous convolutional models and show better or comparable results with other speech enhancement baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑