voice-dataset-lili / README.md
vsisik's picture
update
49a7fca verified
metadata
dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: transcription
      dtype: string
    - name: normalized_transcription
      dtype: string
  splits:
    - name: train
      num_bytes: 102250143.1
      num_examples: 1150
  download_size: 102820339
  dataset_size: 102250143.1
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
language:
  - sk
size_categories:
  - 1K<n<10K
task_categories:
  - automatic-speech-recognition
  - text-to-speech
task_ids: []
tags:
  - speech-recognition
  - tts
  - Speech-To-Text
  - Text-To-Speech

Model card for lili (medium)

  • Language: sk_SK (Slovak, Slovakia)
  • Speakers: 1
  • Quality: medium
  • Samplerate: 22,050Hz

Dataset