ae-mask-distilled-libri-one-sec-cv12/chunk_166

该数据集包含音频和对应的浮点数序列特征。音频特征的采样率为16000Hz，浮点数序列特征用于存储模型输出的logits。数据集仅包含一个训练集，共有2977个样本，总大小为95440015.1841896字节。数据集的下载大小为91238968字节。数据文件路径配置在默认配置中指定。

This dataset contains audio and corresponding sequence of float features. The audio feature has a sampling rate of 16000Hz, and the sequence of float features is used to store the models output logits. The dataset includes only a training set with 2977 samples, totaling 95440015.1841896 bytes in size. The download size of the dataset is 91238968 bytes. The data file paths are specified in the default configuration.

原始信息汇总

数据集概述

特征

• audio:
  • 数据类型: 音频
  • 采样率: 16000
• logits:
  • 数据类型: float32

数据集划分

• train:
  • 样本数量: 2977
  • 数据大小: 95440015.1841896 字节

数据集大小

• 下载大小: 91238968 字节
• 总数据大小: 95440015.1841896 字节

配置

• default:
  • 数据文件路径: data/train-*