metadata
dataset_info:
features:
- name: audio_filename
dtype: string
- name: audio
dtype: audio
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 72372422
num_examples: 490
- name: test
num_bytes: 21209984
num_examples: 140
- name: validation
num_bytes: 10195551
num_examples: 70
download_size: 103778114
dataset_size: 103777957
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*