README.md

metadata

dataset_info:
  features:
    - name: sentence
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
  splits:
    - name: test
      num_bytes: 49708261
      num_examples: 107
    - name: train
      num_bytes: 595749122.6
      num_examples: 1200
  download_size: 577691471
  dataset_size: 645457383.6
configs:
  - config_name: default
    data_files:
      - split: test
        path: data/test-*
      - split: train
        path: data/train-*
license: cc0-1.0
task_categories:
  - automatic-speech-recognition
language:
  - cy