---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 9214214437.921028
    num_examples: 34092
  - name: test
    num_bytes: 1143301837.041485
    num_examples: 4262
  - name: valid
    num_bytes: 1134736508.4274852
    num_examples: 4262
  download_size: 11008882557
  dataset_size: 11492252783.389997
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---