论文标题
与文本无关的扬声器验证的多任务指标学习
Multi-task Metric Learning for Text-independent Speaker Verification
论文作者
论文摘要
在这项工作中，我们介绍了公制学习（ML），以增强与文本无关的说话者验证（SV）的深层嵌入学习。具体而言，深入扬声器嵌入网络经过常规的横熵损失和基于辅助对的ML损耗函数训练。对于辅助ML任务，首先将迷你批次的训练样本排列成对，然后通过自己和相对的相似性选择正面和负对，然后对辅助ML损失进行加权。为了评估所提出的方法，我们在野外（SITW）数据集中的说话者上进行实验。结果证明了该方法的有效性。
In this work, we introduce metric learning (ML) to enhance the deep embedding learning for text-independent speaker verification (SV). Specifically, the deep speaker embedding network is trained with conventional cross entropy loss and auxiliary pair-based ML loss function. For the auxiliary ML task, training samples of a mini-batch are first arranged into pairs, then positive and negative pairs are selected and weighted through their own and relative similarities, and finally the auxiliary ML loss is calculated by the similarity of the selected pairs. To evaluate the proposed method, we conduct experiments on the Speaker in the Wild (SITW) dataset. The results demonstrate the effectiveness of the proposed method.
