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 2011.04092cs.SDcs.NEeess.AS
收藏

频率选通:时频域语音增强的改进卷积神经网络

Frequency Gating: Improved Convolutional Neural Networks for Speech Enhancement in the Time-Frequency Domain

University of Science and Technology of China(中国科学技术大学)

Koen Oostermeijer, Qing Wang, Jun Du

摘要:

传统卷积神经网络(CNN)的优势之一是其固有的平移不变性。然而,对于在时间-频率域中的语音增强的任务,由于在频率方向上缺乏不变性,不能充分利用该属性。在本文中,我们提出通过引入一种方法来弥补这种低效率,我们称之为频率门控,以计算CNN的内核的乘法权重,以使它们依赖于频率。探讨了几种机制:时间选通,其中权重取决于先前时间帧;局部选通,其权重基于单个时间帧和与其相邻的时间帧生成;以及频率选通,其中每个内核被分配独立于输入数据的权重。与自动编码器的神经网络与跳跃连接的实验表明,本地和频率明智的门优于基线,因此是可行的方法来改善基于CNN的语音增强神经网络。此外,一个损失函数的基础上扩展的短时客观可懂度得分(ESTOI),我们表明优于标准均方误差(MSE)损失函数。

英文摘要:

One of the strengths of traditional convolutional neural networks (CNNs) is their inherent translational invariance. However, for the task of speech enhancement in the time-frequency domain, this property cannot be fully exploited due to a lack of invariance in the frequency direction. In this paper we propose to remedy this inefficiency by introducing a method, which we call Frequency Gating, to compute multiplicative weights for the kernels of the CNN in order to make them frequency dependent. Several mechanisms are explored: temporal gating, in which weights are dependent on prior time frames, local gating, whose weights are generated based on a single time frame and the ones adjacent to it, and frequency-wise gating, where each kernel is assigned a weight independent of the input data. Experiments with an autoencoder neural network with skip connections show that both local and frequency-wise gating outperform the baseline and are therefore viable ways to improve CNN-based speech enhancement neural networks. In addition, a loss function based on the extended short-time objective intelligibility score (ESTOI) is introduced, which we show to outperform the standard mean squared error (MSE) loss function.

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。

↑