---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: label
    dtype:
      class_label:
        names:
          '0': sound
          '1': speech
  splits:
  - name: train
    num_bytes: 61697802662.090996
    num_examples: 11331
  - name: eval
    num_bytes: 6548451427.7039995
    num_examples: 1259
  download_size: 31182982123
  dataset_size: 68246254089.795
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
---


experimental dataset - private for now, more details coming soon