README.md

metadata

dataset_info:
  features:
    - name: sentence
      dtype: string
    - name: language
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: prediction
      dtype: string
  splits:
    - name: test
      num_bytes: 954231819.75
      num_examples: 5386
  download_size: 821832215
  dataset_size: 954231819.75
configs:
  - config_name: default
    data_files:
      - split: test
        path: data/test-*