JigsawStack/fleurs_mlt_benchmark

这是一个包含音频数据和对应转录文本的数据集，数据集分为训练集。音频数据包括数组形式的声音数据、路径和采样率。此外，数据集还包含转录文本的原始和清洗版本、性别、语言ID、语言和语言组ID。数据集的总大小为245,207,171字节，下载大小为150,245,639字节，训练集包含145个示例。

This is a dataset containing audio data and corresponding transcription text, split into a training set. The audio data includes array-formatted audio, path, and sampling rate. Additionally, the dataset contains the original and cleaned versions of the transcription text, gender, language ID, language, and language group ID. The total size of the dataset is 245,207,171 bytes, with a download size of 150,245,639 bytes, and the training set contains 145 examples.