---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 572771313.9729787
    num_examples: 1654
  - name: test
    num_bytes: 75052324.80851063
    num_examples: 207
  - name: valid
    num_bytes: 67733011.80851063
    num_examples: 207
  download_size: 678885260
  dataset_size: 715556650.59
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---