metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 805494922.108
num_examples: 2204
- name: validation
num_bytes: 172468787
num_examples: 381
download_size: 972866903
dataset_size: 977963709.108
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*