---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: transcripts
    dtype: string
  splits:
  - name: train
    num_bytes: 2488721141.145171
    num_examples: 12890
  - name: test
    num_bytes: 326389692.0569508
    num_examples: 1612
  - name: valid
    num_bytes: 308662185.54587823
    num_examples: 1611
  download_size: 3086264891
  dataset_size: 3123773018.748
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: valid
    path: data/valid-*
---