---
dataset_info:
  pretty_name: Salan Dataset
  license: cc-by-4.0
  task_categories:
    - automatic-speech-recognition
  language:
    - bdm
    - dje
    - dzg
    - ful
    - gux
    - hau
    - kby
    - taq
  features:
    - name: mp3
      dtype:
        audio:
          decode: false
    - name: txt
      dtype: string
    - name: language_code
      dtype: string
    - name: __key__
      dtype: string
    - name: __url__
      dtype: string
    
  builder_name: webdataset

  data_files:
  - split: bdm
    path:
      - audio/bdm/bdm.tar
  - split: dje
    path:
      - audio/dje/dje.tar
  - split: dzg
    path:
      - audio/dzg/dzg.tar
  - split: ful
    path:
      - audio/ful/ful.tar
  - split: gux
    path:
      - audio/gux/gux.tar
  - split: hau
    path:
      - audio/hau/hau.tar
  - split: kby
    path:
      - audio/kby/kby.tar
  - split: taq
    path:
      - audio/taq/taq.tar

---