README.md

metadata

task_categories:
  - automatic-speech-recognition
language:
  - en
dataset_info:
  features:
    - name: audio_path
      dtype: string
    - name: speaker_type
      dtype: string
    - name: text
      dtype: string
    - name: timestamps
      dtype: string
    - name: c_word
      dtype: string
    - name: audio
      dtype: audio
    - name: id
      dtype: int64
    - name: audio_len_ms
      dtype: float32
    - name: text_len
      dtype: int32
    - name: wps
      dtype: float32
  splits:
    - name: train
      num_bytes: 4280306397.013422
      num_examples: 26644
    - name: validation
      num_bytes: 611426442.9902824
      num_examples: 3806
    - name: test
      num_bytes: 1223174182.06201
      num_examples: 7614
  download_size: 6754080825
  dataset_size: 6114907022.065714
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
      - split: test
        path: data/test-*