audiblehealthai/cough-detection

该数据集包含咳嗽声音和非咳嗽声音的音频文件。音频文件的采样率为16000Hz，每个音频文件都有对应的路径、文件ID、声音类型、起始索引、结束索引、是否为咳嗽、起始秒数、结束秒数和持续时间等属性。数据集分为训练集和验证集，其中训练集包含15972个示例，验证集包含6719个示例。

The dataset includes audio files of cough and non-cough sounds. The audio files have a sampling rate of 16000Hz, and each audio file has corresponding attributes such as path, file ID, sound type, start index, end index, whether it is a cough, start second, end second, and duration seconds. The dataset is divided into training and validation sets, with the training set containing 15972 examples and the validation set containing 6719 examples.