metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 24000
- name: text
dtype: string
- name: source
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 63351657.108391605
num_examples: 257
- name: test
num_bytes: 7079683.891608391
num_examples: 29
download_size: 62970056
dataset_size: 70431341
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*