metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcript_text
dtype: string
- name: language
dtype: string
- name: dialect
dtype: string
splits:
- name: test
num_bytes: 39604703.38076923
num_examples: 70
download_size: 34726271
dataset_size: 39604703.38076923
configs:
- config_name: default
data_files:
- split: test
path: data/test-*