metadata
dataset_info:
features:
- name: text
dtype: string
- name: text_len
dtype: int64
- name: language
dtype: string
- name: speaker_name
dtype: string
- name: duration
dtype: float64
- name: audio
dtype: audio
- name: file_size
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 781818740.114
num_examples: 3189
download_size: 655603537
dataset_size: 781818740.114
configs:
- config_name: default
data_files:
- split: train
path: data/train-*