metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 2332757832.46
num_examples: 6435
- name: test
num_bytes: 224637696.579
num_examples: 2797
- name: validation
num_bytes: 85968232.574
num_examples: 3162
download_size: 2553050674
dataset_size: 2643363761.613
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*