Morocco-MTNRA-Labs/audio-ar

该数据集包含音频、文本、正字法、来源和持续时间等特征。音频特征的采样率为16000 Hz，文本和正字法特征为字符串类型，来源特征为字符串类型，持续时间特征为浮点型。数据集包含一个名为train的分割，共有100,000个样本，总大小约为30.3 GB，下载大小约为30.2 GB。默认配置指向训练数据文件。

The dataset includes features such as Audio, Text, Orthography, Source, and Duration. The Audio feature has a sampling rate of 16000 Hz, Text and Orthography features are of string type, Source feature is of string type, and Duration feature is of float32 type. The dataset contains a single split named train with 100,000 examples, a total size of approximately 30.3 GB, and a download size of about 30.2 GB. The default configuration points to the training data files.