metadata
dataset_info:
features:
- name: audio_array
list: float32
- name: sampling_rate
dtype: int32
- name: text
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 103117938
num_examples: 238
download_size: 69893326
dataset_size: 103117938
configs:
- config_name: default
data_files:
- split: train
path: data/train-*