---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: sampling_rate
    dtype: int64
  - name: transcript
    dtype: string
  - name: lang_spans
    dtype: string
  - name: dominant_lang
    dtype: string
  splits:
  - name: train
    num_bytes: 63843252472.8
    num_examples: 205200
  - name: test
    num_bytes: 7428122064.0
    num_examples: 22800
  download_size: 69543252936
  dataset_size: 71271374536.8
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---