metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 1441207450.3
num_examples: 3420
- name: test
num_bytes: 360037769
num_examples: 856
download_size: 1794863978
dataset_size: 1801245219.3