---
dataset_info:
  features:
  - name: mic
    dtype:
      audio:
        sampling_rate: 16000
  - name: ref
    dtype:
      audio:
        sampling_rate: 16000
  - name: clean
    dtype:
      audio:
        sampling_rate: 16000
  splits:
  - name: train
    num_bytes: 9601406220
    num_examples: 10000
  download_size: 7851570451
  dataset_size: 9601406220
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---