论文标题
DEAAN：用于强大的说话者表示学习的解开嵌入和对抗适应网络
DEAAN: Disentangled Embedding and Adversarial Adaptation Network for Robust Speaker Representation Learning
论文作者
论文摘要
尽管讲话者的验证随着深度神经网络的发展取得了重大的性能提高，但在该领域，域失配仍然是一个具有挑战性的问题。在这项研究中，我们提出了一个新颖的框架，以使扬声器相关和特定于域特异性特征，并仅在与扬声器相关的特征空间上应用域的适应性。使用拆卸可以有效提高适应性性能，而不是直接在未删除域信息的特征空间上进行域适应。具体来说，我们的模型的输入语音来自源和目标域首先被编码为不同的潜在特征空间。对抗域的适应性是在共同的说话者相关的特征空间上进行的，以鼓励域 - 不变性的属性。此外，我们将两个领域的与说话者相关的特定特征和特定领域的特征之间的相互信息最小化，以执行分离。 Voices数据集的实验结果表明，与原始的基于RESNET的系统相比，我们提出的框架可以有效地产生更多的扬声器 - 歧义和域名扬声器表示，而EER相对降低了20.3％。
Despite speaker verification has achieved significant performance improvement with the development of deep neural networks, domain mismatch is still a challenging problem in this field. In this study, we propose a novel framework to disentangle speaker-related and domain-specific features and apply domain adaptation on the speaker-related feature space solely. Instead of performing domain adaptation directly on the feature space where domain information is not removed, using disentanglement can efficiently boost adaptation performance. To be specific, our model's input speech from the source and target domains is first encoded into different latent feature spaces. The adversarial domain adaptation is conducted on the shared speaker-related feature space to encourage the property of domain-invariance. Further, we minimize the mutual information between speaker-related and domain-specific features for both domains to enforce the disentanglement. Experimental results on the VOiCES dataset demonstrate that our proposed framework can effectively generate more speaker-discriminative and domain-invariant speaker representations with a relative 20.3% reduction of EER compared to the original ResNet-based system.
