metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: language
dtype: string
- name: Part
dtype: string
- name: duration
dtype: string
- name: word_count
dtype: string
splits:
- name: train
num_bytes: 5441326324.988
num_examples: 3723
- name: test
num_bytes: 3720399798.528
num_examples: 1584
download_size: 3192665773
dataset_size: 9161726123.515999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*