amuvarma/100-400-raw

该数据集包含300,000个训练样本，每个样本包含一个文本转录（transcript）和六个facodec序列（facodec_0到facodec_5），这些序列的数据类型为int64。数据集的总大小为17,213,260,331字节，下载大小为2,732,616,033字节。数据集仅包含一个训练分割（train）。

This dataset contains 300,000 training samples, each consisting of a text transcript and six facodec sequences (facodec_0 to facodec_5) with a data type of int64. The total size of the dataset is 17,213,260,331 bytes, with a download size of 2,732,616,033 bytes. The dataset includes only a training split (train).