Datasets:

danielroncel
/

dstc2_audios_input_hubert

dataset_info:
  features:
    - name: session_ids
      dtype: string
    - name: path
      dtype: string
    - name: input_values
      sequence: float32
  splits:
    - name: train
      num_bytes: 4900832859
      num_examples: 25516
  download_size: 3317106728
  dataset_size: 4900832859
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*