---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcription
    dtype: string
  - name: locale
    dtype: string
  splits:
  - name: train
    num_bytes: 4025909295.625
    num_examples: 18347
  - name: test
    num_bytes: 573716568.5
    num_examples: 10700
  - name: validation
    num_bytes: 210877557.784
    num_examples: 1304
  download_size: 5366410927
  dataset_size: 4810503421.909
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: validation
    path: data/validation-*
---