---
dataset_info:
  features:
  - name: audios
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 41766736888.684
    num_examples: 1396949
  download_size: 152849897412
  dataset_size: 41766736888.684
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---