---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcription
    dtype: string
  splits:
  - name: train
    num_bytes: 621015746.8685446
    num_examples: 894
  - name: test
    num_bytes: 187135363.08763692
    num_examples: 256
  - name: valid
    num_bytes: 98381996.04381846
    num_examples: 128
  download_size: 775819937
  dataset_size: 906533106.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---