metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: duration
dtype: float64
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 134530727598.282
num_examples: 245093
download_size: 5374089950
dataset_size: 134530727598.282