metadata

license: cc0-1.0
task_categories:
  - conversational
language:
  - 'no'
  - nb
  - nn
pretty_name: NB Samtale — Conversations
size_categories:
  - 1K<n<10K

NB Samtale — Conversations

This dataset contains extracted and cleaned conversations from the NB Samtale corpus. The original is a speech corpus made by the Language Bank at the National Library of Norway. The corpus contains orthographically transcribed speech from podcasts and recordings of live events.