SONYC-UST-V2
收藏arXiv2025-09-30 收录
下载链接：
https://zenodo.org/record/3966543
下载链接
链接失效反馈官方服务：
资源简介：
该数据集名为SONYC，包含了来自纽约市不同地点的61个麦克风录制的18,515个音频片段，这些音频片段被标注为8个粗粒度类别和23个细粒度类别。每个音频片段的时长为10秒，采样率为48kHz。标注信息涵盖了录音的时间、地点以及传感器ID。该数据集的规模为18,515个音频片段，其任务是对声音事件进行检测（SED）。
The dataset named SONYC contains 18,515 audio clips recorded by 61 microphones deployed at various locations across New York City. All audio clips are annotated with 8 coarse-grained categories and 23 fine-grained categories. Each audio clip has a duration of 10 seconds and a sampling rate of 48 kHz. The annotation metadata includes the recording timestamp, geographic location, and sensor ID. The total number of audio clips in this dataset is 18,515, and its dedicated task is sound event detection (SED).
提供机构：
SONYC
