metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 35906758
num_examples: 160
- name: test
num_bytes: 4342414
num_examples: 20
- name: valid
num_bytes: 4821908
num_examples: 20
download_size: 45069287
dataset_size: 45071080
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*