metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: text
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 532184873
num_examples: 427
download_size: 491496703
dataset_size: 532184873
configs:
- config_name: default
data_files:
- split: train
path: data/train-*