librilight_medium / README.md
theodorr's picture
Upload dataset (part 00003-of-00004)
91c78e0 verified
metadata
dataset_info:
  features:
    - name: text
      dtype: string
    - name: align
      sequence: string
    - name: speaker
      dtype: string
    - name: book
      dtype: string
    - name: audio_token
      sequence:
        sequence:
          sequence: int64
    - name: align_token
      sequence:
        sequence: string
    - name: text_token
      sequence: int64
    - name: audio_duration
      dtype: float64
  splits:
    - name: train
      num_bytes: 88217493434
      num_examples: 763341
  download_size: 15126644868
  dataset_size: 88217493434
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*