---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 108047558909.34956
    num_examples: 880971
  - name: test
    num_bytes: 13957212384.713346
    num_examples: 110122
  - name: valid
    num_bytes: 14083017814.714104
    num_examples: 110121
  download_size: 130808669625
  dataset_size: 136087789108.77701
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---