metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: upload_id
dtype: string
- name: version
dtype: int64
- name: segment_id
dtype: int64
- name: start
dtype: float64
- name: end
dtype: float64
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 36083841094
num_examples: 41054
download_size: 36671422573
dataset_size: 36083841094
configs:
- config_name: default
data_files:
- split: train
path: data/train-*