---
dataset_info:
  config_name: train
  features:
  - name: id
    dtype: string
  - name: text
    dtype: string
  - name: input_features
    dtype: image
  - name: condition_on_prev
    sequence: int64
  - name: whisper_transcript
    sequence: int64
  splits:
  - name: train
    num_bytes: 4009681656693.0
    num_examples: 2607550
  download_size: 2515115685452
  dataset_size: 4009681656693.0
configs:
- config_name: train
  data_files:
  - split: train
    path: train/train-*
---