Teroye/common_voice_diarization

该数据集包含音频数据及其相关元信息，如分段、客户端ID、索引、文本、性别和年龄等。数据集分为训练集、测试集和验证集，分别包含2108、2277和1831个样本。总下载大小为16313436545字节，数据集总大小为19513961916.624字节。

The dataset includes multiple features such as audio, segments, client IDs, indices, texts, genders, and ages. It is divided into train, test, and validation splits, each with specified byte sizes and number of examples. The total download size of the dataset is 16313436545 bytes, and the actual size is 19513961916.624 bytes. The dataset configuration includes the default config and corresponding data file paths.

原始信息汇总

数据集概述

数据集特征

• audio: 音频数据
• segments: 整数序列
• client_ids: 字符串序列
• indices: 字符串序列
• texts: 字符串序列
• genders: 字符串序列
• ages: 字符串序列

数据集划分

• train:
  • 样本数量: 2108
  • 字节数: 10262238367.056
• test:
  • 样本数量: 2277
  • 字节数: 5000772792.878
• validation:
  • 样本数量: 1831
  • 字节数: 4250950756.69

数据集大小

• 下载大小: 16313436545 字节
• 数据集总大小: 19513961916.624 字节

配置

• config_name: default
  • data_files:
    • train: data/train-*
    • test: data/test-*
    • validation: data/validation-*