maxseats/aihub-464-preprocessed-680GB-set-30
收藏Hugging Face2024-07-02 更新2024-07-06 收录
下载链接:
https://hf-mirror.com/datasets/maxseats/aihub-464-preprocessed-680GB-set-30
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含三个分割:训练集、测试集和验证集。每个分割的数据文件路径和大小都有详细说明。数据集的特征包括音频、标签和输入特征,音频的采样率为16000Hz,标签为字符串类型,输入特征为浮点数序列。训练集包含49072个示例,测试集和验证集各包含6134个示例。整个数据集的下载大小为22406815947字节,总大小为69577712206.0字节。
The dataset contains three splits: train, test, and valid. The data file paths and sizes for each split are detailed. The features of the dataset include audio, labels, and input features, with the audio sampled at 16000Hz, labels as strings, and input features as sequences of float32. The train set contains 49,072 examples, while the test and valid sets each contain 6,134 examples. The total download size of the dataset is 22,406,815,947 bytes, and the total dataset size is 69,577,712,206.0 bytes.
提供机构:
maxseats
原始信息汇总
数据集概述
数据集配置
- 配置名称: default
- 数据文件:
- 训练集: data/train-*
- 测试集: data/test-*
- 验证集: data/valid-*
数据集信息
- 特征:
- 音频:
- 采样率: 16000
- 标签: string
- 输入特征:
- 序列: float32
- 音频:
数据集分割
- 训练集:
- 字节数: 55662169764.8
- 样本数: 49072
- 测试集:
- 字节数: 6957771220.6
- 样本数: 6134
- 验证集:
- 字节数: 6957771220.6
- 样本数: 6134
数据集大小
- 下载大小: 22406815947
- 数据集总大小: 69577712206.0



