---
dataset_info:
  features:
  - name: voice
    dtype: string
  - name: text
    dtype: string
  - name: caption
    dtype: string
  - name: language
    dtype: string
  - name: audio
    dtype: audio
  splits:
  - name: train
    num_bytes: 30006400616
    num_examples: 28783
  download_size: 16998686310
  dataset_size: 30006400616
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---