---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: speaker_id
    dtype: string
  - name: question
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: train
    num_bytes: 3287914474.0
    num_examples: 10000
  download_size: 3153212824
  dataset_size: 3287914474.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---