MTS团队 @ AutoMin 2021：现有摘要方法概览及与无监督摘要技术的比较
Team MTS @ AutoMin 2021: An Overview of Existing Summarization Approaches and Comparison to Unsupervised Summarization Techniques
AI总结：
本文针对AutoMin 2021自动会议纪要任务，分析现有摘要方法并提出基于聚类的无监督摘要技术，在真实录音流水线上优于预训练模型。
摘要：
由于全球疫情，通过视频或音频会议进行的远程通信变得比以往任何时候都更加普及。这些事件因此推动了面向口语的自动会议纪要系统的发展，进而催生了AutoMin 2021挑战赛。本文展示了MTS团队在参加Automatic Minutes挑战赛期间所开展研究的结果。具体而言，本文分析了现有的文本和语音摘要方法，提出了一种基于聚类的无监督摘要技术，并提供了一个包含适配的自动语音识别模块、能够在真实录音上运行的流水线。所提出的无监督技术在自动会议纪要任务上优于预训练摘要模型，在开发集上的Rouge 1、Rouge 2和Rouge L值分别为0.21、0.02和0.2，在测试集上的Rouge 1、Rouge 2、Rouge L、Adequacy、Grammatical correctness和Fluency值分别为0.180、0.035、0.098、1.857、2.304和1.911。
英文摘要：
Remote communication through video or audio conferences has become more popular than ever because of the worldwide pandemic. These events, therefore, have provoked the development of systems for automatic minuting of spoken language leading to AutoMin 2021 challenge. The following paper illustrates the results of the research that team MTS has carried out while participating in the Automatic Minutes challenge. In particular, in this paper we analyze existing approaches to text and speech summarization, propose an unsupervised summarization technique based on clustering and provide a pipeline that includes an adapted automatic speech recognition block able to run on real-life recordings. The proposed unsupervised technique outperforms pre-trained summarization models on the automatic minuting task with Rouge 1, Rouge 2 and Rouge L values of 0.21, 0.02 and 0.2 on the dev set, with Rouge 1, Rouge 2, Rouge L, Adequacy, Grammatical correctness and Fluency values of 0.180, 0.035, 0.098, 1.857, 2.304, 1.911 on the test set accordingly
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
