metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcripts
dtype: string
splits:
- name: train
num_bytes: 7854422.290076336
num_examples: 104
- name: test
num_bytes: 1043347.9236641221
num_examples: 14
- name: valid
num_bytes: 1272508.786259542
num_examples: 13
download_size: 8236632
dataset_size: 10170279
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*