ocisd4/english_asr

该数据集包含两个配置：librispeech和tedlium。每个配置都包含音频和消息两个特征，其中音频特征的数据类型为audio，消息特征是一个列表，包含内容和角色两个字段，数据类型均为string。每个配置还包含训练集的分割信息，包括字节数、示例数、下载大小和数据集大小。此外，还提供了每个配置的数据文件路径。

The dataset includes two configurations: librispeech and tedlium. Each configuration contains two features: audio and messages. The audio feature has a data type of audio, and the messages feature is a list containing content and role fields, both with a data type of string. Each configuration also includes split information for the training set, including the number of bytes, number of examples, download size, and dataset size. Additionally, the data file paths for each configuration are provided.