arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.03512eess.AScs.LGcs.SD
收藏

基于信噪比特征和多样训练数据的鲁棒DNN语音增强

SNR-Based Features and Diverse Training Data for Robust DNN-Based Speech Enhancement

Robert Rehr, Timo Gerkmann

摘要:

在本文中,我们解决了基于深度神经网络(DNN)的语音增强在训练数据大小和多样性有限的情况下对不可见噪声条件的泛化。为了获得更多的见解,我们分析了泛化方面:(1)训练数据的大小和多样性,(2)不同的网络架构,以及(3)选择的特征。为了解决(1),我们在Hu噪声语料库(有限大小)、CHiME 3噪声语料库(有限多样性)上训练网络,并且还提出了基于免费可用的声音收集的大而多样的数据集。为了解决(2),我们比较了全连接前馈和长短期记忆(LSTM)架构。为了解决(3),我们比较了三个输入特征,即量化的噪声周期图,噪声感知训练(NAT)和提出的基于信噪比(SNR)的噪声感知训练(SNR-NAT)。我们证实,丰富的训练数据和改进的网络架构有助于DNN的推广。此外,我们通过实验结果和使用t分布随机邻居嵌入(t-SNE)的分析表明,即使使用简单的网络架构,并且仅在小数据集上训练时,所提出的SNR-NAT特征也会在不可见的噪声中产生鲁棒和水平独立的结果,这是本文的关键贡献。

英文摘要:

In this paper, we address the generalization of deep neural network (DNN) based speech enhancement to unseen noise conditions for the case that training data is limited in size and diversity. To gain more insights, we analyze the generalization with respect to (1) the size and diversity of the training data, (2) different network architectures, and (3) the chosen features. To address (1), we train networks on the Hu noise corpus (limited size), the CHiME 3 noise corpus (limited diversity) and also propose a large and diverse dataset collected based on freely available sounds. To address (2), we compare a fully-connected feed-forward and a long short-term memory (LSTM) architecture. To address (3), we compare three input features, namely logarithmized noisy periodograms, noise aware training (NAT) and the proposed signal-to-noise ratio (SNR) based noise aware training (SNR-NAT). We confirm that rich training data and improved network architectures help DNNs to generalize. Furthermore, we show via experimental results and an analysis using t-distributed stochastic neighbor embedding (t-SNE) that the proposed SNR-NAT features yield robust and level independent results in unseen noise even with simple network architectures and when trained on only small datasets, which is the key contribution of this paper.

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。

↑