README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: duration
      dtype: float32
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 36993063181.52
      num_examples: 123971
  download_size: 16692928156
  dataset_size: 36993063181.52
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
task_categories:
  - automatic-speech-recognition
language:
  - ar