---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: dataset
    dtype: string
  - name: task
    dtype: string
  - name: split
    dtype: string
  - name: duration
    dtype: float64
  - name: wave_filename
    dtype: string
  - name: answer_formatted
    dtype: string
  splits:
  - name: train
    num_bytes: 74099956359.172
    num_examples: 234603
  download_size: 75084829681
  dataset_size: 74099956359.172
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---