rahul7star/hindi-speech-dataset

该数据集是一个包含音频和对应文本转录的多模态数据集。它被划分为训练集、验证集和测试集，分别包含145,152个、239个和418个样本。数据集总大小约为18GB，下载大小约为17GB。

This dataset is a multimodal dataset containing audio and corresponding text transcripts. It is divided into training, validation, and test sets, containing 145,152, 239, and 418 samples respectively. The total size of the dataset is approximately 18GB, and the download size is about 17GB.