DCASE 2021 Task 5: Few-shot Bioacoustic Event Detection Development Set

Name: DCASE 2021 Task 5: Few-shot Bioacoustic Event Detection Development Set
Creator: Zenodo
Published: 2021-02-20 14:57:10
License: 暂无描述

Zenodo2021-02-20 更新2026-05-25 收录

下载链接：

https://zenodo.org/record/4543504

下载链接

链接失效反馈

官方服务：

资源简介：

General Description The development set for task 5 of DCASE 2021 "Few-shot Bioacoustic Event Detection" consists of 19 audio files acquired from different bioacoustic sources. The dataset is split into training and validation Sets. Multi-class annotations are provided for the training set with positive (POS), negative (NEG) and unkwown (UNK) values for each class. UNK indicates uncertainty about a class. Single-class (class of interest) annotations are provided for the validation set, with events marked as positive (POS) or unkwown (UNK) provided for the class of interest. Folder Structure Development_Set.zip |_Development_Set/ |__Training_Set/ |___BV/ |____*.wav |____*.csv |___HT/ |____*.wav |____*.csv |___JD/ |____*.wav |____*.csv |___MT/ |____*.wav |____*.csv |__Validation_Set/ |___HV/ |____*.wav |____*.csv |___PB/ |____*.wav |____*.csv Development_Set_Audio.zip has the same structure but contains only the *.wav files. Development_Set_Annotations.zip has the same structure but contains only the *.csv files Dataset statistics Some statistics on this dataset are as follows, split between training and validation set and their sub-folders: ----------------------------------------------------- TRAINING SET ----------------------------------------------------- Number of audio recordings | 11 Total duration | 14 hours and 20 mins Total classes (excl. UNK) | 19 Total events (excl. UNK) | 4,686 ----------------------------------------------------- TRAINING SET/BV ----------------------------------------------------- Number of audio recordings | 5 Total duration | 10 hours Total classes (excl. UNK) | 11 Total events (excl. UNK) | 2,662 Sampling rate | 24,000 Hz ----------------------------------------------------- TRAINING SET/HT ----------------------------------------------------- Number of audio recordings | 3 Total duration | 3 hours Total classes (excl. UNK) | 3 Total events (excl. UNK) | 435 Sampling rate | 6,000 Hz ----------------------------------------------------- TRAINING SET/JD ----------------------------------------------------- Number of audio recordings | 1 Total duration | 10 mins Total classes (excl. UNK) | 1 Total events (excl. UNK) | 355 Sampling rate | 22,050 Hz ----------------------------------------------------- TRAINING SET/MT ----------------------------------------------------- Number of audio recordings | 2 Total duration | 1 hour and 10 mins Total classes (excl. UNK) | 4 Total events (excl. UNK) | 1,234 Sampling rate | 8,000 Hz ----------------------------------------------------- ----------------------------------------------------- VALIDATION SET ----------------------------------------------------- Number of audio recordings | 8 Total duration | 5 hours Total classes (excl. UNK) | 4 Total events (excl. UNK) | 310 ----------------------------------------------------- VALIDATION SET/HV ----------------------------------------------------- Number of audio recordings | 2 Total duration | 2 hours Total classes (excl. UNK) | 2 Total events (excl. UNK) | 50 Sampling rate | 6,000 Hz ----------------------------------------------------- VALIDATION SET/PB ----------------------------------------------------- Number of audio recordings | 6 Total duration | 3 hours Total classes (excl. UNK) | 2 Total events (excl. UNK) | 260 Sampling rate | 44,100 Hz ----------------------------------------------------- Annotation structure Each line of the annotation csv represents an event in the audio file. The column descriptions are as follows: TRAINING SET --------------------- Audiofilename, Starttime, Endtime, CLASS_1, CLASS_2, ...CLASS_N VALIDATION SET --------------------- Audiofilename, Starttime, Endtime, Q Open Access This dataset is available under a Creative Commons Attribution 4.0 International (CC BY 4.0) license. Contact info Please send any feedback or questions to: Veronica Morfi: g.v.morfi@qmul.ac.uk

**数据集概述** 本数据集为DCASE 2021“少样本生物声学事件检测（Few-shot Bioacoustic Event Detection）”任务5的开发集，包含19个采集自不同生物声学源的音频文件。数据集被划分为训练集（Training Set）与验证集（Validation Set）。训练集配备多类别标注，每个类别包含正样本（POS）、负样本（NEG）与未知样本（UNK，Unknown），其中UNK表示类别归属存在不确定性。验证集仅提供目标类别的单类别标注，将目标类别的事件标记为正样本（POS）或未知样本（UNK）。 **文件夹结构** *Development_Set.zip* |_Development_Set/ |__Training_Set/ |___BV/ |____*.wav |____*.csv |___HT/ |____*.wav |____*.csv |___JD/ |____*.wav |____*.csv |___MT/ |____*.wav |____*.csv |__Validation_Set/ |___HV/ |____*.wav |____*.csv |___PB/ |____*.wav |____*.csv 其中*Development_Set_Audio.zip*具有完全一致的目录结构，但仅包含*.wav音频文件；*Development_Set_Annotations.zip*同样具有相同目录结构，仅包含*.csv标注文件。 **数据集统计信息** 本数据集的统计信息按训练集、验证集及其子文件夹划分如下： ----------------------------------------------------- TRAINING SET（训练集） ----------------------------------------------------- 音频录音总数 | 11 总时长 | 14小时20分钟总类别数（不含UNK） | 19 总事件数（不含UNK） | 4686 ----------------------------------------------------- TRAINING SET/BV（训练集/BV） ----------------------------------------------------- 音频录音总数 | 5 总时长 | 10小时总类别数（不含UNK） | 11 总事件数（不含UNK） | 2662 采样率 | 24000 Hz ----------------------------------------------------- TRAINING SET/HT（训练集/HT） ----------------------------------------------------- 音频录音总数 | 3 总时长 | 3小时总类别数（不含UNK） | 3 总事件数（不含UNK） | 435 采样率 | 6000 Hz ----------------------------------------------------- TRAINING SET/JD（训练集/JD） ----------------------------------------------------- 音频录音总数 | 1 总时长 | 10分钟总类别数（不含UNK） | 1 总事件数（不含UNK） | 355 采样率 | 22050 Hz ----------------------------------------------------- TRAINING SET/MT（训练集/MT） ----------------------------------------------------- 音频录音总数 | 2 总时长 | 1小时10分钟总类别数（不含UNK） | 4 总事件数（不含UNK） | 1234 采样率 | 8000 Hz ----------------------------------------------------- ----------------------------------------------------- VALIDATION SET（验证集） ----------------------------------------------------- 音频录音总数 | 8 总时长 | 5小时总类别数（不含UNK） | 4 总事件数（不含UNK） | 310 ----------------------------------------------------- VALIDATION SET/HV（验证集/HV） ----------------------------------------------------- 音频录音总数 | 2 总时长 | 2小时总类别数（不含UNK） | 2 总事件数（不含UNK） | 50 采样率 | 6000 Hz ----------------------------------------------------- VALIDATION SET/PB（验证集/PB） ----------------------------------------------------- 音频录音总数 | 6 总时长 | 3小时总类别数（不含UNK） | 2 总事件数（不含UNK） | 260 采样率 | 44100 Hz ----------------------------------------------------- **标注文件结构** 标注CSV文件的每一行对应音频文件中的一个事件，各列说明如下： ### 训练集标注CSV --------------------- 音频文件名（Audiofilename）、开始时间（Starttime）、结束时间（Endtime）、类别1（CLASS_1）、类别2（CLASS_2）……类别N（CLASS_N） ### 验证集标注CSV --------------------- 音频文件名（Audiofilename）、开始时间（Starttime）、结束时间（Endtime）、Q **开放获取说明** 本数据集采用知识共享署名4.0国际许可（Creative Commons Attribution 4.0 International，CC BY 4.0）协议发布。 **联系方式** 如有任何反馈或疑问，请联系： Veronica Morfi：g.v.morfi@qmul.ac.uk

提供机构：

Zenodo

创建时间：

2021-02-19