metadata
dataset_info:
features:
- name: id
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 99244661.84969094
num_examples: 2254
- name: validation
num_bytes: 5277982.608486789
num_examples: 119
- name: test
num_bytes: 5508087.277822258
num_examples: 125
download_size: 104425436
dataset_size: 110030731.73599999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*