README.md

metadata

dataset_info:
  features:
    - name: audio_id
      dtype: string
    - name: voice_creator_id
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: audio_duration
      dtype: float64
    - name: gender
      dtype: string
    - name: age_group
      dtype: string
    - name: category
      dtype: string
    - name: subcategory
      dtype: string
    - name: location
      dtype: string
    - name: transcription
      dtype: string
    - name: transcription_with_tags
      dtype: string
  splits:
    - name: train
      num_bytes: 43583964712.972
      num_examples: 64541
    - name: validation
      num_bytes: 2592262640.12
      num_examples: 3989
  download_size: 43972117303
  dataset_size: 46176227353.092
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: validation
        path: data/validation-*
license: cc-by-4.0
task_categories:
  - automatic-speech-recognition
language:
  - sw
size_categories:
  - 10K<n<100K