joycewu/common_voice_16_1_zh_TW_pseudo_labelled_non_concat2
收藏Hugging Face2024-11-22 更新2024-12-14 收录
下载链接：
https://hf-mirror.com/datasets/joycewu/common_voice_16_1_zh_TW_pseudo_labelled_non_concat2
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含多个特征，如client_id、path、audio、sentence等，其中audio特征包含采样率为16000的音频数据。数据集分为train、validation和test三个部分，每个部分都有相应的字节大小和示例数量。数据集的下载大小为411678601字节，总大小为433059122.643字节。配置文件中指定了每个部分的数据文件路径。
The dataset configuration name is zh-TW, containing multiple features such as client_id, path, audio, etc., with the audio feature having a sampling rate of 16000. The dataset is divided into training, validation, and test sets, containing 7101, 4976, and 4976 samples respectively. The total download size of the dataset is 411678601 bytes, and the dataset size is 433059122.643 bytes.
提供机构：
joycewu
