README.md

metadata

dataset_info:
  features:
    - name: relative_path
      dtype: string
    - name: section
      dtype: string
    - name: filename
      dtype: string
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 5926501
      num_examples: 1730
  download_size: 3557313
  dataset_size: 5926501
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: odc-by
task_categories:
  - text-generation
language:
  - en
size_categories:
  - 1K<n<10K

The Sarcasm Scrolls

Paragraph-segmented audio transcriptions of various satirical media.