metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: caption
dtype: string
- name: transcript
dtype: string
- name: '5000'
dtype: string
- name: '10000'
dtype: string
- name: '15000'
dtype: string
- name: '20000'
dtype: string
- name: '25000'
dtype: string
- name: '30000'
dtype: string
- name: '35000'
dtype: string
- name: '40000'
dtype: string
- name: '45000'
dtype: string
- name: '50000'
dtype: string
splits:
- name: train
num_bytes: 2231036387
num_examples: 5000
download_size: 2182548588
dataset_size: 2231036387
configs:
- config_name: default
data_files:
- split: train
path: data/train-*