README.md

metadata

dataset_info:
  features:
    - name: speaker_id
      dtype: int64
    - name: audio
      dtype: audio
    - name: gender
      dtype: string
    - name: sentence
      dtype: string
    - name: environment
      dtype: string
    - name: duration
      dtype: float64
    - name: generator
      dtype: string
    - name: spoofed
      dtype: int64
  splits:
    - name: train
      num_bytes: 18753218681.92
      num_examples: 46188
    - name: test
      num_bytes: 2092305658.076
      num_examples: 5141
  download_size: 17966169741
  dataset_size: 20845524339.996
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
license: cc-by-nc-sa-3.0
language:
  - th