---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: question
    dtype: string
  - name: reference_answer
    dtype: string
  - name: reference_reasoning
    dtype: 'null'
  - name: generated_answer
    dtype: string
  - name: generated_reasoning
    dtype: string
  - name: audio
    dtype:
      audio:
        decode: false
  splits:
  - name: train
    num_bytes: 78753121.0
    num_examples: 50
  download_size: 78072149
  dataset_size: 78753121.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---