---
dataset_info:
  features:
  - name: conversation_id
    dtype: int64
  - name: turn_id
    dtype: int64
  - name: speaker_id
    dtype: int64
  - name: text
    dtype: string
  - name: audio
    dtype:
      audio:
        sampling_rate: 24000
  splits:
  - name: train
    num_bytes: 1891953.7734404576
    num_examples: 12
  download_size: 1128945
  dataset_size: 1891953.7734404576
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---