---
dataset_info:
  features:
  - name: audio
    dtype: audio
  - name: speech_id
    dtype: string
  - name: speaker_id
    dtype: string
  - name: transcript
    dtype: string
  splits:
  - name: train
    num_bytes: 1585742876.38
    num_examples: 3446
  - name: test
    num_bytes: 276840471.0
    num_examples: 680
  download_size: 1668931471
  dataset_size: 1862583347.38
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---