README.md

metadata

dataset_info:
  features:
    - name: id
      dtype: string
    - name: clean
      dtype:
        audio:
          sampling_rate: 16000
    - name: noisy
      dtype:
        audio:
          sampling_rate: 16000
  splits:
    - name: train
      num_bytes: 2373596219.84
      num_examples: 11572
    - name: test
      num_bytes: 132699198
      num_examples: 824
  download_size: 2283241065
  dataset_size: 2506295417.84
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
license: cc-by-4.0
task_categories:
  - audio-to-audio
language:
  - en
pretty_name: ' VoiceBank+DEMAND'
paperswithcode_id: demand