README.md

metadata

license: cc-by-4.0
task_categories:
  - audio-to-audio
  - automatic-speech-recognition
  - audio-classification
  - text-to-speech
task_ids:
  - speaker-identification
size_categories:
  - 10K<n<100K
source_datasets: []
language:
  - fr
multilinguality:
  - monolingual
language_creators:
  - crowdsourced
  - expert-generated
annotations_creators:
  - expert-generated
pretty_name: VibraVox
configs:
  - config_name: ASR_Reference_microphone
    data_files:
      - split: train
        path: ASR_reference_mic/clean/train/train_ASR_ref_microphone.tsv
      - split: val
        path: ASR_reference_mic/clean/val/val_ASR_ref_microphone.tsv
      - split: test
        path: ASR_reference_mic/clean/test/test_ASR_ref_microphone.tsv
  - config_name: ASR_Laryngophone
    features:
      - name: file_name
        dtype: string
      - name: audio
        dtype: audio
      - name: transcription
        dtype: string
      - name: sensor_id
        dtype: string
      - name: speaker_id
        dtype: int64
      - name: gender
        dtype: string
      - name: type
        dtype: string
      - name: gender
        dtype: string
      - name: split
        dtype: string
    data_files:
      - split: train
        path: train_ASR_laryngophone.tsv
      - split: val
        path: val_ASR_laryngophone.tsv
      - split: test
        path: test_ASR_laryngophone.tsv