espnet/ml_superb_hf
收藏Hugging Face2024-12-28 更新2024-12-14 收录
下载链接：
https://hf-mirror.com/datasets/espnet/ml_superb_hf
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含音频和文本数据，音频采样率为16000Hz。数据集分为开发集（dev）和训练集（train），分别包含23892和122890个样本。开发集文件大小为1886862795.984字节，训练集文件大小为12913911474.48字节。总下载大小为14483202989字节，数据集总大小为14800774270.463999字节。数据集的特征包括id、language、audio和text。
The dataset includes four features: id, language, audio, and text. The sampling rate for the audio feature is 16000. The dataset is divided into two splits: dev and train, containing 23892 and 124546 samples respectively. The total download size of the dataset is 14632846462 bytes, and the total dataset size is 14974731910.656 bytes. The dataset configuration is named default, which includes data file paths for the dev and train splits.
提供机构：
espnet
