论文标题
交叉专心汇总扬声器验证
Cross attentive pooling for speaker verification
论文作者
论文摘要
本文的目的是与文本无关的扬声器验证，其中发音来自“野外”视频，并且可能包含无关的信号。虽然说话者验证自然是一个成对的问题，但现有的产生扬声器嵌入的方法是实例的。在本文中，我们提出了跨专心池（CAP），该合并（CAP）利用参考对中的上下文信息来生成包含对成对匹配问题的最判别信息的话语级嵌入。实验是在Voxceleb数据集上进行的，其中我们的方法的表现优于可比的合并策略。
The goal of this paper is text-independent speaker verification where utterances come from 'in the wild' videos and may contain irrelevant signal. While speaker verification is naturally a pair-wise problem, existing methods to produce the speaker embeddings are instance-wise. In this paper, we propose Cross Attentive Pooling (CAP) that utilizes the context information across the reference-query pair to generate utterance-level embeddings that contain the most discriminative information for the pair-wise matching problem. Experiments are performed on the VoxCeleb dataset in which our method outperforms comparable pooling strategies.
