用于文本无关说话人确认的端到端可训练自注意浅层网络
End-to-End Trainable Self-Attentive Shallow Network for Text-Independent Speaker Verification
摘要：
广义端到端（GE2E）模型因其可扩展性和通用性而被广泛应用于说话人确认（SV）领域。然而，基于GE2E的长短期记忆（LSTM）有两个局限性：第一，GE2E的嵌入遭受消失梯度，这导致非常长的输入序列的性能下降。其次，话语没有被表示为适当固定的维度向量。在本文中，为了克服上述问题，我们提出了一种新的SV框架，端到端可训练的自注意浅层网络（SASN），结合了一个时间延迟神经网络（TDNN）和一个自注意池机制的基础上自注意的x向量系统在话语嵌入阶段。我们证明，该模型是高效的，并提供更准确的说话人确认比GE2E。对于VCTK数据集，仅小于GE2E的一半大小，所提出的模型在EER（等误差率）、DCF（检测代价函数）和AUC（曲线下面积）方面分别显示出比GE2E显著的性能改进，分别为约63%、67%和85%。值得注意的是，当输入长度变得更长时，所提出的模型的DCF分数改善是GE2E的约17倍。
英文摘要：
Generalized end-to-end (GE2E) model is widely used in speaker verification (SV) fields due to its expandability and generality regardless of specific languages. However, the long-short term memory (LSTM) based on GE2E has two limitations: First, the embedding of GE2E suffers from vanishing gradient, which leads to performance degradation for very long input sequences. Secondly, utterances are not represented as a properly fixed dimensional vector. In this paper, to overcome issues mentioned above, we propose a novel framework for SV, end-to-end trainable self-attentive shallow network (SASN), incorporating a time-delay neural network (TDNN) and a self-attentive pooling mechanism based on the self-attentive x-vector system during an utterance embedding phase. We demonstrate that the proposed model is highly efficient, and provides more accurate speaker verification than GE2E. For VCTK dataset, with just less than half the size of GE2E, the proposed model showed significant performance improvement over GE2E of about 63%, 67%, and 85% in EER (Equal error rate), DCF (Detection cost function), and AUC (Area under the curve), respectively. Notably, when the input length becomes longer, the DCF score improvement of the proposed model is about 17 times greater than that of GE2E.
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。
