metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 2526495389.0143003
num_examples: 3726
- name: test
num_bytes: 635180006.9010239
num_examples: 929
download_size: 3072085903
dataset_size: 3161675395.915324
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*