README.md

metadata

language:
  - ko
size_categories:
  - 100K<n<1M
task_categories:
  - feature-extraction
pretty_name: Korea_address
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
      - split: valid
        path: data/valid-*
dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: transcripts
      dtype: string
  splits:
    - name: train
      num_bytes: 2581554.5517241377
      num_examples: 23
    - name: test
      num_bytes: 544091.448275862
      num_examples: 6
    - name: valid
      num_bytes: 2581554.5517241377
      num_examples: 23
  download_size: 4642711
  dataset_size: 5707200.551724138