README.md

metadata

dataset_info:
  features:
    - name: client_id
      dtype: string
    - name: path
      dtype: audio
    - name: sentence
      dtype: string
    - name: up_votes
      dtype: int64
    - name: down_votes
      dtype: int64
    - name: age
      dtype: string
    - name: gender
      dtype: string
    - name: accents
      dtype: string
    - name: variant
      dtype: float64
    - name: locale
      dtype: string
    - name: segment
      dtype: string
  splits:
    - name: train
      num_bytes: 1061971115.28
      num_examples: 26328
    - name: test
      num_bytes: 377748044.084
      num_examples: 10196
    - name: validated
      num_bytes: 5852414660.504
      num_examples: 158417
    - name: other
      num_bytes: 642660551.92
      num_examples: 12585
    - name: invalidated
      num_bytes: 467995235.5
      num_examples: 9795
  download_size: 7829612580
  dataset_size: 8402789607.288
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: validated
        path: data/validated-*
      - split: other
        path: data/other-*
      - split: invalidated
        path: data/invalidated-*