cff-version: 1.2.0
message: "If you use this dataset, please cite it as below."
title: "Chinese Debate Documents: ASR-transcribed competitive debate corpus"
abstract: >-
  A collection of ASR-generated transcripts of Mandarin Chinese university
  debates, primarily drawn from the 第十一届新国辩 (11th New International Debate
  Open) and related collegiate tournaments. Transcripts are speaker-segmented
  with timestamps and (where available) carry topic, round, and team metadata
  parsed from the original audio filenames.
type: dataset
authors:
  - family-names: "DBWBD"
    given-names: ""
license: MIT
repository: "https://huggingface.co/datasets/DBWBD/Chinese_Debate_Documents"
url: "https://huggingface.co/datasets/DBWBD/Chinese_Debate_Documents"
keywords:
  - chinese
  - mandarin
  - debate
  - transcripts
  - asr
  - nlp
date-released: 2025-07-29