---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: audio
    dtype: string
  - name: num_audio_samples
    dtype: int64
  - name: sampling_rate
    dtype: int64
  - name: conversation
    list:
    - name: from
      dtype: string
    - name: value
      dtype: string
  - name: language
    dtype: string
  splits:
  - name: train
    num_bytes: 967115652
    num_examples: 2419047
  - name: test
    num_bytes: 403790
    num_examples: 1000
  download_size: 409911870
  dataset_size: 967519442
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---