burkimbia/speech-dataset-clean

该数据集包含多个特征，包括文本、音频、持续时间、说话者、来源、性别等。音频数据未解码，其他特征包括是否保留、是否重复、剪辑状态、静音比例、字符每秒、单词计数和检测到的语言。数据集分为训练集，包含87,929个样本，总大小为15,881,513,055字节。

The dataset includes multiple features such as text, audio (not decoded), duration, speakers, source, gender, and others like keep status, duplicate status, clipped status, silence ratio, characters per second, word count, and detected language. It is split into a training set with 87,929 examples and a total size of 15,881,513,055 bytes.