README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
          decode: false
    - name: text
      dtype: string
    - name: speaker_name
      dtype: string
  splits:
    - name: train
      num_bytes: 6778898640
      num_examples: 30504
    - name: test
      num_bytes: 750280661
      num_examples: 3390
  download_size: 7125544218
  dataset_size: 7529179301
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*
language:
  - ro