AudioMNIST_subset / README.md
robello2's picture
Upload dataset
8e57720 verified
metadata
dataset_info:
  features:
    - name: speaker_id
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: digit
      dtype:
        class_label:
          names:
            '0': '0'
            '1': '1'
            '2': '2'
            '3': '3'
            '4': '4'
            '5': '5'
            '6': '6'
            '7': '7'
            '8': '8'
            '9': '9'
    - name: gender
      dtype:
        class_label:
          names:
            '0': male
            '1': female
    - name: accent
      dtype: string
    - name: age
      dtype: int64
    - name: native_speaker
      dtype: bool
    - name: origin
      dtype: string
  splits:
    - name: train
      num_bytes: 118078722
      num_examples: 2000
  download_size: 84765507
  dataset_size: 118078722
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*