metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: speaker
dtype: int64
- name: start
dtype: float64
- name: end
dtype: float64
- name: audio_duration
dtype: float64
splits:
- name: train
num_bytes: 49428388057
num_examples: 406138
download_size: 56287699493
dataset_size: 49428388057