---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: round
    dtype: int64
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: answer_token
    dtype: 'null'
  - name: question_audio
    dtype:
      audio:
        sampling_rate: 16000
  splits:
  - name: train
    num_bytes: 89701799063.916
    num_examples: 421092
  download_size: 99324815822
  dataset_size: 89701799063.916
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---