jq/sunbird_speech

---
dataset_info:
- config_name: ach
features:
- name: ids
dtype: string
- name: texts
dtype: string
- name: audios
sequence: float32
- name: audio_languages
dtype: string
- name: are_studio
dtype: bool
- name: speaker_ids
dtype: string
- name: sample_rates
dtype: int64
splits:
- name: train
num_bytes: 3820274
num_examples: 10
- name: dev
num_bytes: 3580883
num_examples: 10
- name: test
num_bytes: 3742241
num_examples: 10
download_size: 5375242
dataset_size: 11143398
- config_name: lug
features:
- name: ids
dtype: string
- name: texts
dtype: string
- name: audios
sequence: float32
- name: audio_languages
dtype: string
- name: are_studio
dtype: bool
- name: speaker_ids
dtype: string
- name: sample_rates
dtype: int64
splits:
- name: train
num_bytes: 5196706
num_examples: 10
- name: dev
num_bytes: 3370989
num_examples: 10
- name: test
num_bytes: 2899936
num_examples: 10
download_size: 5732034
dataset_size: 11467631
configs:
- config_name: ach
data_files:
- split: train
path: ach/train-*
- split: dev
path: ach/dev-*
- split: test
path: ach/test-*
- config_name: lug
data_files:
- split: train
path: lug/train-*
- split: dev
path: lug/dev-*
- split: test
path: lug/test-*
---

原始信息汇总

数据集概述

配置 ach

• 特征:
  • ids: 类型为字符串
  • texts: 类型为字符串
  • audios: 类型为 float32 序列
  • audio_languages: 类型为字符串
  • are_studio: 类型为布尔值
  • speaker_ids: 类型为字符串
  • sample_rates: 类型为 int64
• 分割:
  • train: 字节数为 3820274，样本数为 10
  • dev: 字节数为 3580883，样本数为 10
  • test: 字节数为 3742241，样本数为 10
• 下载大小: 5375242 字节
• 数据集大小: 11143398 字节
• 数据文件:
  • train: ach/train-*
  • dev: ach/dev-*
  • test: ach/test-*

配置 lug

• 特征:
  • ids: 类型为字符串
  • texts: 类型为字符串
  • audios: 类型为 float32 序列
  • audio_languages: 类型为字符串
  • are_studio: 类型为布尔值
  • speaker_ids: 类型为字符串
  • sample_rates: 类型为 int64
• 分割:
  • train: 字节数为 5196706，样本数为 10
  • dev: 字节数为 3370989，样本数为 10
  • test: 字节数为 2899936，样本数为 10
• 下载大小: 5732034 字节
• 数据集大小: 11467631 字节
• 数据文件:
  • train: lug/train-*
  • dev: lug/dev-*
  • test: lug/test-*