amuvarma/dualcodec-face-tokenised-joined

该数据集包含视频相关的信息，如视频ID、唯一标识符、视频片段、令牌、语义代码、转录文本以及多个类别标签（cb_0到cb_6）。每个视频片段可能对应多个令牌和语义代码，以及多个类别标签。数据集分为训练集，共有超过428万示例，数据集大小约为27.7GB。The dataset includes video-related information such as video ID, unique identifier, video segment, tokens, semantic codes, transcription text, and multiple category labels (cb_0 to cb_6). Each video segment may correspond to multiple tokens, semantic codes, and category labels. The dataset is split into a training set with over 4.28 million examples, and the dataset size is approximately 27.7GB.