A dataset of dialogue data drawn from the BBC Radio 4 Moral Maze discussion show
Clone the Git repo or download the dataset (zipped or tarred) and get started working with the dialogue data. Each dialogue is stored in JSON file that has two major parts:
Contains a JSON document consisting of 'speaker' and 'utterance' pairs
Contains a JSON document that describes the data such as when and where it was collected, how it was processed, who worked on the data, and any errors that have been discovered. The meta-data section is described in the /schema/datafile.json file which is stored the projects Git repo.
This project is maintained by @siwells
If you are having trouble or want to contribute then contact simon.wells@abdn.ac.uk.