jingyaogong-ai/17.12.2025

该数据集包含音频、转录文本和视频ID三个主要特征。音频数据的采样率为16000Hz。数据集分为训练集一个部分，包含39649个样本，总大小为37008168409字节。下载大小为33797080516字节。The dataset includes three main features: audio, transcript, and video_id. The audio data has a sampling rate of 16000Hz. The dataset is divided into a training split with 39,649 examples, totaling 37,008,168,409 bytes. The download size is 33,797,080,516 bytes.