---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: score
    dtype: float64
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 5346976563.35
    num_examples: 41433
  download_size: 5145514547
  dataset_size: 5346976563.35
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---