---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcription
    dtype: string
  - name: source
    dtype: string
  - name: length_s
    dtype: float64
  - name: cleaned_transcription
    dtype: string
  splits:
  - name: train
    num_bytes: 13028688980.0
    num_examples: 88712
  - name: test
    num_bytes: 330934008.084
    num_examples: 1708
  download_size: 12907433193
  dataset_size: 13359622988.084
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---