metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 7950015165.473
num_examples: 5211
- name: test
num_bytes: 1345940254.064
num_examples: 1104
- name: validation
num_bytes: 275255625
num_examples: 361
download_size: 9266054751
dataset_size: 9571211044.536999