metadata

dataset_info:
  features:
    - name: client_id
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: sentence_id
      dtype: string
    - name: sentence
      dtype: string
    - name: up_votes
      dtype: int64
    - name: down_votes
      dtype: int64
    - name: age
      dtype: string
    - name: gender
      dtype: string
    - name: accents
      dtype: string
    - name: variant
      dtype: float64
    - name: locale
      dtype: string
  splits:
    - name: validated_without_test
      num_bytes: 1225104973.706
      num_examples: 49402
    - name: test
      num_bytes: 144958326.717
      num_examples: 5013
  download_size: 1320268405
  dataset_size: 1370063300.423
configs:
  - config_name: default
    data_files:
      - split: validated_without_test
        path: data/validated_without_test-*
      - split: test
        path: data/test-*
license: cc0-1.0
task_categories:
  - automatic-speech-recognition
language:
  - zh
pretty_name: Common Voice Corpus 19.0 Chinese (Taiwan)

Common Voice Corpus 19.0 Chinese (Taiwan)

The test set is the same as the original test set, while validated_without_test includes all validated examples except those with sentence IDs that appear in the test set.

validated_without_test has about 50,000 examples in total, equivalent to approximately 44 hours, and is intended for use as the training set.
test has about 5,000 examples, which is approximately 5 hours.