arda-argmax/voxpopuli_en_pseudo_labelled
收藏Hugging Face2025-04-09 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/arda-argmax/voxpopuli_en_pseudo_labelled
下载链接
链接失效反馈官方服务：
资源简介：
这是一个包含音频数据和对应文本的英文数据集，采样率为16000Hz。数据集分为训练集、验证集和测试集，分别包含79611、760和739个示例。数据集特征包括音频ID、归一化文本、是否基于前一个条件以及耳语转录文本。
This is an English dataset containing audio data and corresponding text, with a sampling rate of 16000Hz. The dataset is divided into training, validation, and test sets, containing 79611, 760, and 739 examples respectively. The features of the dataset include audio ID, normalized text, condition based on the previous one, and whisper transcript text.
提供机构：
arda-argmax
