metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 3215899.2
num_examples: 24
- name: test
num_bytes: 372035.4
num_examples: 3
- name: valid
num_bytes: 463315.4
num_examples: 3
download_size: 3560654
dataset_size: 4051250
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*