README.md

metadata

dataset_info:
  features:
    - name: path
      dtype: string
    - name: audio
      dtype:
        audio:
          sampling_rate: 16000
    - name: sentence
      dtype: string
  splits:
    - name: train
      num_bytes: 901857303.92
      num_examples: 4972
  download_size: 965382804
  dataset_size: 901857303.92

Dataset Card for "mgb2_audios_transcriptions_non_overlap"

More Information needed