asahi417/seamless-align-enA-koA.tokenized.encodec

该数据集包含多个子集（subset_1到subset_46），每个子集包含英文和韩文音频的ID、LASER评分以及音频的token序列。数据集主要用于训练模型，每个子集的训练集大小和示例数量不同。

This dataset contains multiple subsets (subset_1 to subset_46), each of which includes IDs, LASER scores, and token sequences for English and Korean audio. The dataset is primarily used for training models, with varying sizes and numbers of examples in each subsets training set.