ae-mask-distilled-libri-one-sec-cv12/chunk_120

该数据集包含音频和对应的logits序列。音频数据的采样率为16000Hz，logits是一个浮点数序列。数据集包含一个训练分割，共有10613个样本，总大小为340243493.7085248字节。数据集的下载大小为322556304字节。数据文件路径为data/train-*。

The dataset contains audio and corresponding logits sequences. The audio data has a sampling rate of 16000Hz, and the logits are a sequence of float32 numbers. The dataset includes a training split with 10613 examples, totaling 340243493.7085248 bytes in size. The download size of the dataset is 322556304 bytes. The data files are located at data/train-*.

原始信息汇总

数据集概述

特征

• audio:
  • dtype:
    • sampling_rate: 16000
• logits:
  • sequence: float32

数据分割

• train:
  • num_bytes: 340243493.7085248
  • num_examples: 10613

数据集大小

• download_size: 322556304
• dataset_size: 340243493.7085248

配置

• config_name: default
  • data_files:
    • split: train
    • path: data/train-*