---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 598099904.5376791
    num_examples: 1934
  - name: test
    num_bytes: 74221544.33416046
    num_examples: 242
  - name: valid
    num_bytes: 74699082.33416046
    num_examples: 242
  download_size: 665620175
  dataset_size: 747020531.206
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---