---
dataset_info:
  features:
  - name: file_name
    dtype: string
  - name: transcription
    dtype: string
  - name: transcription_normalised
    dtype: string
  - name: audio
    dtype: audio
  splits:
  - name: train
    num_bytes: 4983072167.73
    num_examples: 20978
  download_size: 3741291896
  dataset_size: 4983072167.73
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---