OpenSpeechHub/mls_eng_10k-llama-1b-tokenized
收藏Hugging Face2025-07-04 更新2025-07-05 收录
下载链接：
https://hf-mirror.com/datasets/OpenSpeechHub/mls_eng_10k-llama-1b-tokenized
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含音频文件的相关信息，如原始路径、起始时间、结束时间、音频时长、说话者ID、书籍ID等。同时，它还提供了用于模型训练的输入ID、标签和注意力掩码等信息。数据集被划分为训练集，大小为约38GB，共有约242万个示例。数据集的下载大小约为12GB。
This dataset contains information about audio files, such as original path, begin time, end time, audio duration, speaker ID, book ID, etc. It also provides input IDs, labels, and attention masks for model training. The dataset is split into a training set, which is about 38GB in size and contains approximately 2.42 million examples. The download size of the dataset is about 12GB.
提供机构：
OpenSpeechHub
