metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: source
dtype: string
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 6437450059.725216
num_examples: 25835
- name: eval
num_bytes: 49835102.87478395
num_examples: 200
download_size: 5866183197
dataset_size: 6487285162.599999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*