metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: type
dtype: string
- name: fluency
dtype: int64
splits:
- name: train
num_bytes: 482850514.9
num_examples: 2710
download_size: 470794329
dataset_size: 482850514.9
configs:
- config_name: default
data_files:
- split: train
path: data/train-*