metadata
dataset_info:
features:
- name: audio
sequence:
sequence: float16
- name: seq_name
dtype: string
splits:
- name: train
num_bytes: 15552086400
num_examples: 2880
- name: validation
num_bytes: 6561036450
num_examples: 1215
download_size: 16246620427
dataset_size: 22113122850
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*