First Shared Task on Automatic Minuting at Interspeech 2021最新文献

Overview of the First Shared Task on Automatic Minuting (AutoMin) at Interspeech 2021 Interspeech 2021第一个自动记录(AutoMin)共享任务概述

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-1

Tirthankar Ghosal, Ondrej Bojar, Muskaan Singh, A. Nedoluzhko

引用次数: 22

Team JU_PAD @ AutoMin 2021: MoM Generation from Multiparty Meeting Transcript 团队JU_PAD @ AutoMin 2021:多方会议记录中的MoM一代

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-5

Sarthak Pan, Palash Nandi, Dipankar Das

{"title":"Team JU_PAD @ AutoMin 2021: MoM Generation from Multiparty Meeting Transcript","authors":"Sarthak Pan, Palash Nandi, Dipankar Das","doi":"10.21437/automin.2021-5","DOIUrl":"https://doi.org/10.21437/automin.2021-5","url":null,"abstract":"Use of online meeting platforms for long multi-party discus-sion is gradually increasing and generation of Minutes of Meeting (MoM) is crucial for subsequent events. MOM records all key issues, possible solutions, decisions and actions taken dur-ing the meeting. Hence the importance of minuting cannot be overemphasized in a time when a significant number of meet-ings take place in the virtual space. Automatic generation of MoM can potentially save up to 80% of time while revisiting. In this paper, we present an abstractive approach for automatic generation of meeting minutes. It aims to deal with problems like the nature of spoken text, length of transcripts and lack of document structure and conversation fillers. The system is evaluated on a test dataset. The evaluation score is calculated by both manual and automatic systems. Text summarization metrics ROUGE-1, ROUGE-2, ROUGE-L [1] are used for automated scoring and metrics Adequacy, Grammatical Correctness, Fluency are used for manual scoring. The proposed model achieved 0.221, 0.046, 0,125 for ROUGE-1, ROUGE-2 , ROUGE-L respectively in automated evaluation and 3.5/5, 3/5, 3/5 for Adequacy, Grammatical Correctness, Fluency respectively in manual evaluation.","PeriodicalId":186820,"journal":{"name":"First Shared Task on Automatic Minuting at Interspeech 2021","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124292559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Team Symantlytical @ AutoMin 2021: Generating Readable Minutes with GPT-2 and BERT-based Automatic Minuting Approach 赛门铁克团队@ AutoMin 2021:使用GPT-2和基于bert的自动记录方法生成可读的记录

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-8

Amitesh Garg, Muskaan Singh

{"title":"Team Symantlytical @ AutoMin 2021: Generating Readable Minutes with GPT-2 and BERT-based Automatic Minuting Approach","authors":"Amitesh Garg, Muskaan Singh","doi":"10.21437/automin.2021-8","DOIUrl":"https://doi.org/10.21437/automin.2021-8","url":null,"abstract":"This paper describes our participation system run to Automatic Minuting @ Interspeech 2021 1 . The task was motivated towards generating automatic minutes. We make a initial step towards, namely Main Task A , Task B and Task C . The main task A, was to automatically create minutes from multiparty meeting transcripts, while task B to identify whether the minute belongs to the transcript and task C. GPT-2[1]. The shared task, consist-ing of three subtasks, required to produce, contrast and scruti-nize the meeting minutes. The process of automating minuting is considered to be one of the most challenging tasks in natural language processing and sequence-to-sequence transforma-tion. It involves testing the semantic meaningfulness, readability and reasonable adequacy of the Minutes produced in the system. In the proposed work, we have developed a system using pre-trained language models in order to generate dialogue summaries or minutes. The designed methodology considers cov-erage, adequacy and readability to produce the best utilizable summary of a meeting transcript with any length. Our evaluation results in subtask A achieve a score of 11% R-L which by far is the most challenging than subtask as it required systems to generate the rational minutes of the given meeting transcripts.","PeriodicalId":186820,"journal":{"name":"First Shared Task on Automatic Minuting at Interspeech 2021","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126682838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Team The Turing TESTament @ AutoMin 2021: A Pipeline based Approach to Generate Meeting Minutes Using TOPSIS 团队图灵遗嘱@ AutoMin 2021:使用TOPSIS生成会议纪要的基于管道的方法

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-9

Umang Sharma, Muskaan Singh, Harpreet Singh

引用次数: 2

Team ABC @ AutoMin 2021: Generating Readable Minutes with a BART-based Automatic Minuting Approach 团队ABC @ AutoMin 2021:使用基于bart的自动记录方法生成可读的会议记录

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-2

Kartik Shinde, Nidhir Bhavsar, Aakash Bhatnagar, Tirthankar Ghosal

{"title":"Team ABC @ AutoMin 2021: Generating Readable Minutes with a BART-based Automatic Minuting Approach","authors":"Kartik Shinde, Nidhir Bhavsar, Aakash Bhatnagar, Tirthankar Ghosal","doi":"10.21437/automin.2021-2","DOIUrl":"https://doi.org/10.21437/automin.2021-2","url":null,"abstract":"This paper documents the approach of our Team ABC for the First Shared Task on Automatic Minuting (AutoMin) at Interspeech 2021. This challenge’s primary task (Task A) was to generate meeting minutes from multi-party meeting proceedings automatically. For this purpose, we develop an automatic minuting pipeline where we leverage a denoising autoencoder for pretraining sequence-to-sequence models and ﬁne-tune it on a large-scale abstractive dialogue summarization dataset to summarize meeting transcripts. Speciﬁcally, we use a BART model and train it on the SAMSum dialogue summarization dataset. Our pipeline ﬁrst splits the given transcript into blocks of smaller conversations, eliminates redundancies with a specially-crafted rule-based algorithm, summarizes the conversation blocks, retrieves the block-wise summaries, cleans, structures, and ﬁnally integrates the summaries to produce the meeting minutes. Our proposed system performs the best in several evaluation metrics (automatic, human) in the AutoMin shared task. We use certain text similarity metrics for the subsidiary tasks to determine whether a given transcript-minute pair corresponds to the same meeting (Task B) and if a given pair of meeting minutes belong to the same meeting (Task C). However, our simple machine-learning-based approach did not perform well in addressing the objective of the subsidiary tasks in the challenge. We publicly release our system codes here 1 .","PeriodicalId":186820,"journal":{"name":"First Shared Task on Automatic Minuting at Interspeech 2021","volume":"266 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123106490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 12

Team UEDIN @ AutoMin 2021: Creating Minutes by Learning to Filter an Extracted Summary 团队UEDIN @ AutoMin 2021:通过学习过滤提取的摘要来创建会议记录

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-10

Philip Williams, B. Haddow

引用次数: 2

Team Zoom @ AutoMin 2021: Cross-domain Pretraining for Automatic Minuting 团队缩放@ AutoMin 2021:自动记录的跨域预训练

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-11

Felix Schneider, Sebastian Stüker, V. Parthasarathy

引用次数: 1

Team MTS @ AutoMin 2021: An Overview of Existing Summarization Approaches and Comparison to Unsupervised Summarization Techniques 团队MTS @ AutoMin 2021:现有摘要方法概述及与无监督摘要技术的比较

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-7

O. Iakovenko, A. Andreeva, Anna Lapidus, Liana Mikaelyan

引用次数: 2

Team AutoMinuters @ AutoMin 2021: Leveraging state-of-the-art Text Summarization model to Generate Minutes using Transfer Learning 团队AutoMinuters @ AutoMin 2021:利用最先进的文本摘要模型使用迁移学习生成会议记录

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-3

Parth Mahajan, Muskaan Singh, Harpreet Singh

{"title":"Team AutoMinuters @ AutoMin 2021: Leveraging state-of-the-art Text Summarization model to Generate Minutes using Transfer Learning","authors":"Parth Mahajan, Muskaan Singh, Harpreet Singh","doi":"10.21437/automin.2021-3","DOIUrl":"https://doi.org/10.21437/automin.2021-3","url":null,"abstract":"This paper presents our submission for the first shared task of automatic minuting (AutoMin@Interspeech 2021). The shared task consists of one main task generate minutes from the given meeting transcript. For this challenge, we leveraged state-of-art text summarization models to generate minutes using the transfer learning approach. We also provide an empirical analysis of our proposed method with other text summarization approaches. We evaluate our system submission quantitatively with 33% BERTscore and 11.6 % ROUGE L, which is rela-tively higher than the average submission in the shared task. Along with the qualitative evaluation, we also vouch for quantitative assessment, where we achieve (2.32, 2.64, 2.52) scores out of five for adequacy, grammatical correctness, and fluency. For the other two tasks, we use Jaccard and cosine text similarity metrics for a given transcript-minute pair corresponding to the same meeting (Task B) and if a given pair of meeting minutes belong to the same meeting (Task C). However, our simple approach yielded 94.8 % (task B) and 92.3% (task C), clearly outperforming most submissions in the challenge. We make our codebase release here https://github. com/mahajanparth19/Automin_Submission .","PeriodicalId":186820,"journal":{"name":"First Shared Task on Automatic Minuting at Interspeech 2021","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127112779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

Team Matus and Francesco @ AutoMin 2021: Towards Neural Summarization of Meetings 团队Matus和Francesco @ AutoMin 2021:朝着会议的神经总结

First Shared Task on Automatic Minuting at Interspeech 2021 Pub Date : 2021-09-04 DOI: 10.21437/automin.2021-6

Matús Zilinec, F. Re

{"title":"Team Matus and Francesco @ AutoMin 2021: Towards Neural Summarization of Meetings","authors":"Matús Zilinec, F. Re","doi":"10.21437/automin.2021-6","DOIUrl":"https://doi.org/10.21437/automin.2021-6","url":null,"abstract":"As online meetings are becoming increasingly ubiquitous, there is an increasing demand to record the main outcomes of these meetings for future reference. Automatic summarization of meetings is a challenging, yet relatively unex-plored natural language processing task with a wide range of potential applications. This paper describes our submission to the First Shared Task on Automatic Minuting at Interspeech 2021. In contrast to previous research focused on the summarization of narrated documents, we examine the specifics of bullet-point spoken language summarization on the AutoMin dataset of online meetings in English. Furthermore, we investigate whether existing abstractive summarization systems can be transferred to this new domain. In this regard, we develop a minuting pipeline based on the state-of-the-art PEGASUS summarization model. This includes pre-processing of conversational data, few-shot transfer learning using reference minutes generated by human annotators, filtering and post-processing of the resulting candidate summaries into a suitable bullet-point minutes format. We conclude by evaluating the completeness and shortening aspects of our system, and discuss its limitations and potential future research directions.","PeriodicalId":186820,"journal":{"name":"First Shared Task on Automatic Minuting at Interspeech 2021","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122643558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2