metadata
dataset_info:
features:
- name: id
dtype: string
- name: audio
dtype: audio
- name: text
dtype: string
- name: speaker
dtype: string
- name: language
dtype: string
- name: talkRoleType
dtype: string
- name: talkRoleID
dtype: string
splits:
- name: train
num_bytes: 75999629400.54
num_examples: 109630
download_size: 60275801105
dataset_size: 75999629400.54
configs:
- config_name: default
data_files:
- split: train
path: data/train-*