---
dataset_info:
  features:
  - name: uuid
    dtype: string
  - name: audio
    dtype: audio
  - name: orig_sentence
    dtype: string
  - name: sentence
    dtype: string
  - name: is_retranscribe
    dtype: bool
  - name: transcriber
    dtype: int64
  - name: extra_data
    struct:
    - name: attributes
      struct:
      - name: retranscribe
        dtype: bool
    - name: duration
      dtype: float64
    - name: foreign_language
      dtype: bool
    - name: max_logprob
      dtype: float64
    - name: multiple_speakers
      dtype: bool
    - name: noisy
      dtype: bool
    - name: orig_text
      dtype: string
    - name: skipped
      dtype: bool
    - name: stats
      struct:
      - name: numFullPlays
        dtype: int64
      - name: numPauseClicks
        dtype: int64
      - name: numPlayClicks
        dtype: int64
      - name: referer
        dtype: string
      - name: time
        dtype: float64
      - name: user_agent
        dtype: string
    - name: text
      dtype: string
    - name: too_long
      dtype: bool
    - name: unintelligible
      dtype: bool
  splits:
  - name: train
    num_bytes: 20949729485.454
    num_examples: 203827
  - name: test
    num_bytes: 2702727511.152
    num_examples: 20711
  download_size: 19950659245
  dataset_size: 23652456996.606
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---