---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: original_id
    dtype: string
  - name: question_refined_wav
    dtype: audio
  - name: assistant_a_wav
    dtype: audio
  - name: assistant_b_wav
    dtype: audio
  - name: winner_content
    dtype: string
  - name: winner_style
    dtype: string
  - name: style
    dtype: string
  - name: question_original
    dtype: string
  - name: question_refined
    dtype: string
  - name: assistant_a
    dtype: string
  - name: assistant_b
    dtype: string
  - name: model_text_a
    dtype: string
  - name: model_text_b
    dtype: string
  splits:
  - name: train
    num_bytes: 86601574.0
    num_examples: 20
  download_size: 85454876
  dataset_size: 86601574.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---