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.
