The automated transcription of spoken language, and meetings, in particular, is becoming more widespread as automatic speech recognition systems are becoming more accurate. This trend has significantly accelerated since the outbreak of the COVID-19 pandemic, which led to a major increase in the number of online meetings. However, the transcription of spoken language has not received much attention from the NLP community compared to documents and other forms of written language. In this paper, we study a variation of the summarization problem over the transcription of spoken language: given a transcribed meeting, and an action item (i.e., a commitment or request to perform a task), our goal is to generate a coherent and self-contained rephrasing of the action item. To this end, we compiled a novel dataset of annotated meeting transcripts, including human rephrasing of action items. We use state-of-the-art supervised text generation techniques and establish a strong baseline based on BART and UniLM (two pretrained transformer models). Due to the nature of natural speech, language is often broken and incomplete and the task is shown to be harder than an analogous task over email data. Particularly, we show that the baseline models can be greatly improved once models are provided with additional information. We compare two approaches: one incorporating features extracted by coreference resolution. Additional annotations are used to train an auxiliary model to detect the relevant context in the text. Based on the systematic human evaluation, our best models exhibit near human-level rephrasing capability on a constrained subset of the problem.<\/p>\n","protected":false},"excerpt":{"rendered":"
The automated transcription of spoken language, and meetings, in particular, is becoming more widespread as automatic speech recognition systems are becoming more accurate. This trend has significantly accelerated since the outbreak of the COVID-19 pandemic, which led to a major increase in the number of online meetings. However, the transcription of spoken language has not […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13556,13545],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[246694,246691,246808],"msr-conference":[259084],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-790886","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-human-language-technologies","msr-locale-en_us","msr-field-of-study-artificial-intelligence","msr-field-of-study-computer-science","msr-field-of-study-natural-language-processing"],"msr_publishername":"Association for Computational Linguistics","msr_edition":"","msr_affiliation":"","msr_published_date":"2021-8-1","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"doi","viewUrl":"false","id":"false","title":"10.18653\/V1\/2021.FINDINGS-ACL.253","label_id":"243106","label":0},{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/2021.aclweb.org\/program\/accept\/#911","label_id":"243109","label":0},{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/aclanthology.org\/2021.findings-acl.253\/","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[],"msr-author-ordering":[{"type":"guest","value":"amir-cohen","user_id":792344,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=amir-cohen"},{"type":"edited_text","value":"Amir Kantor (amkantor)","user_id":40153,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Amir Kantor (amkantor)"},{"type":"guest","value":"sagi-hilleli","user_id":792350,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=sagi-hilleli"},{"type":"user_nicename","value":"Eyal Kolman","user_id":41006,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Eyal Kolman"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[644373,916890],"msr_project":[918261,790874],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":918261,"post_title":"Work & well-being","post_name":"work-well-being","post_type":"msr-project","post_date":"2023-10-25 20:54:47","post_modified":"2023-10-25 20:54:50","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/work-well-being\/","post_excerpt":"The COVID-19 pandemic changed the lives of people around the world at home and at work, with effects lasting beyond the lockdowns. Along with Microsoft\u2019s New Future of Work Report 2022, this research looks at the impact of remote work on the personal productivity and well-being of Microsoft\u2019s own employees, the effects of remote work on cross-group collaboration, how working from home affects work relationships, how the pandemic has affected farmers and farmworkers and exploring…","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/918261"}]}},{"ID":790874,"post_title":"Meeting Intelligence: Task Rephrasing","post_name":"meeting-intelligence-task-rephrasing","post_type":"msr-project","post_date":"2021-10-31 04:21:18","post_modified":"2024-09-10 08:56:32","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/meeting-intelligence-task-rephrasing\/","post_excerpt":"Exploratory project to 'decontextualize' tasks and to-do items identified in meeting transcriptions, and rewrite each of them in a single sentence to appear in a separate to-do list. We build upon pretrained seq2seq transformer models, and modern techniques such as multitask learning, and touch upon unresolved issues such as automated metrics to evaluate NLG models. In our teem, we build machine learning technology to detect tasks in emails, and explore the usage of this technology…","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/790874"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/790886"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/790886\/revisions"}],"predecessor-version":[{"id":790904,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/790886\/revisions\/790904"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=790886"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=790886"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=790886"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=790886"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=790886"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=790886"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=790886"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=790886"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=790886"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=790886"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=790886"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=790886"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=790886"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=790886"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=790886"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=790886"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}