使用预训练音频表征学习检测新颖和细粒度声学序列（英文版）

英文标题：Learning to Detect Novel and Fine-Grained Acoustic Sequences Using Pretrained Audio Representations中文摘要：该论文调查了用于少样本声音事件检测的预训练音频表示。他们开发了适合此任务的预训练嵌入并评估其在 AudioSet 上的通用性和在现实世界声学序列构造的任务上的效用。英文摘要：This work investigates pretrained audio representations for few shot SoundEvent Detection. We specifically address the task of few shot detection ofnovel acoustic sequences, or sound events with semantically meaningful temporalstructure, without assuming access to non-target audio. We develop procedure