---
dataset_info:
  features:
  - name: noise_type
    dtype: string
  - name: dirty_audio
    dtype:
      audio:
        sampling_rate: 44100
  - name: generated_audio
    dtype:
      audio:
        sampling_rate: 44100
  - name: clean_audio
    dtype:
      audio:
        sampling_rate: 44100
  splits:
  - name: train
    num_bytes: 18432061.0
    num_examples: 19
  download_size: 18016200
  dataset_size: 18432061.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---