Speaker verification is to judge the similarity of two unknown voices in an open set, where the ideal speaker embedding should be able to condense discriminant information into a compact utterance-level representation that has small intra-speaker distances and large inter-speaker distances.We propose a novel model named Voice Transformer(VOT) for speaker verification. The model consists of multiple parallel Transformers, and the outputs of these Transformers are adaptively combined. Deeply-Fused Semantic Memory Network(DFSMN)is integrated into the attention parts of these Transformers to capture long-distance information and enhance the local dependencies. Statistical pooling layers are incorporated to enhance overall performance without significantly increasing the number of parameters. We propose a new loss function called Additive Angular Margin Focal Loss(AAMF) to address the hard sample mining issue.We evaluate the proposed approach on the VoxCeleb1 and CN-Celeb2 datasets. The experimental results demonstrate that VOT achieves state-of-the-art results, outperforming nearly all existing models. The code is available on GitHub.

翻译：说话人验证是在开放集中判断两段未知语音的相似度，理想的说话人嵌入应能将判别信息压缩成紧凑的语句级表示，使类内距离小、类间距离大。我们提出了一种名为Voice Transformer（VOT）的新型说话人验证模型。该模型由多个并行的Transformer组成，这些Transformer的输出被自适应地组合。在Transformer的注意力部分集成了深度融合语义记忆网络（DFSMN），以捕获长距离信息并增强局部依赖关系。通过融入统计池化层，可在不显著增加参数数量的情况下提升整体性能。我们提出了一种新的损失函数——加性角度边缘焦点损失（AAMF），以解决难样本挖掘问题。在VoxCeleb1和CN-Celeb2数据集上评估了所提方法。实验结果表明，VOT取得了最先进的结果，几乎超越了所有现有模型。代码已开源在GitHub上。