metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: label
dtype: int64
- name: folder
dtype: string
splits:
- name: train
num_bytes: 4153595614.392
num_examples: 78032
- name: test
num_bytes: 345377106.544
num_examples: 6538
- name: validation
num_bytes: 860187438.572
num_examples: 15868
download_size: 9669808840
dataset_size: 5359160159.507999
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*