基于注意力池化的说话人表征
Speaker Characterization by means of Attention Pooling
AI总结：
本文将 Double Multi-Head Self-Attention 池化架构从说话人识别扩展到情感识别、性别分类和 COVID-19 检测等说话人表征任务，并取得优异实验结果。
摘要：
用于说话人验证的最先进深度学习系统通常基于说话人嵌入提取器。这些架构通常由特征提取器前端和池化层组成，以将可变长度话语编码为固定长度的说话人向量。作者近期提出在说话人识别中使用 Double Multi-Head Self-Attention（双重多头自注意力）池化，并将其置于基于 CNN 的前端与一组全连接层之间。结果表明，这是一种出色的方法，能够高效选择前端从语音信号中捕获的最相关特征。本文展示了将该架构适配到其他不同说话人表征任务（如情感识别、性别分类和 COVID-19 检测）所取得的出色实验结果。
英文摘要：
State-of-the-art Deep Learning systems for speaker verification are commonly based on speaker embedding extractors. These architectures are usually composed of a feature extractor front-end together with a pooling layer to encode variable-length utterances into fixed-length speaker vectors. The authors have recently proposed the use of a Double Multi-Head Self-Attention pooling for speaker recognition, placed between a CNN-based front-end and a set of fully connected layers. This has shown to be an excellent approach to efficiently select the most relevant features captured by the front-end from the speech signal. In this paper we show excellent experimental results by adapting this architecture to other different speaker characterization tasks, such as emotion recognition, sex classification and COVID-19 detection.
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。
