metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 413909061.51593626
num_examples: 401
- name: test
num_bytes: 53241352.14541832
num_examples: 51
- name: valid
num_bytes: 47452477.33864542
num_examples: 50
download_size: 495237595
dataset_size: 514602891
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*