README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: sentence
      dtype: string
  splits:
    - name: train
      num_bytes: 28876422583.75
      num_examples: 225722
    - name: test
      num_bytes: 5351760214.5
      num_examples: 41660
  download_size: 34101234883
  dataset_size: 34228182798.25
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: test
        path: data/test-*

Dhivehi Synthetic Audio

Automatic Speech Recognition, Text-to-Speech