---
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
---