metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: duration
dtype: float64
- name: semi-label
dtype: string
- name: corrected-label
dtype: string
splits:
- name: train
num_bytes: 9803819887.31
num_examples: 33282
- name: test
num_bytes: 1710294968.725
num_examples: 5775
download_size: 11503817356
dataset_size: 11514114856.035
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*