[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-517119-105":59,"doc-detail-517119-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","the-shefeld-wargame-corpus-day-two-and-day-three-interspeech-2016-distant-speech-recognition-corpus","The Shefﬁeld Wargame Corpus-Day Two and Day Three - Interspeech 2016 Distant Speech Recognition Corpus","","Improving distant speech recognition performance enables speech technology for everyday home use. Conventional methods enhance signals before recognition, often via beamforming, but few real-world recordings support multichannel experimentation, especially for deep neural approaches that require large training sets with speaker location annotation. This paper introduces an extended freely available English corpus with moving speakers, diverse microphone arrays, and ground-truth location tracking, extending Shefﬁeld Wargames Corpus 2013 with additional fully annotated hours, female speech, blog-based LM data, and a Kaldi baseline.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/the-shefeld-wargame-corpus-day-two-and-day-three-interspeech-2016-distant-speech-recognition-corpus/517119/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/the-shefeld-wargame-corpus-day-two-and-day-three-interspeech-2016-distant-speech-recognition-corpus/517119.png","ImageObject",300,407,{"name":92,"@type":93},"Blitz","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-13","2026-10-07",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",12,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does the paper address in distant speech recognition?","Question",{"text":113,"@type":114},"The paper targets the performance gap between close-talking microphones and distant microphones, where existing approaches and especially deep-learning beamforming methods are limited by scarce real-world multichannel data with location information.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"What is new about the corpus introduced in this work?",{"text":118,"@type":114},"It provides a freely available extended English corpus recorded in natural settings with moving speakers, diverse microphone arrays, and ground-truth speaker location tracking.",{"name":120,"@type":111,"acceptedAnswer":121},"How is the dataset organized and what additional resources are provided?",{"text":122,"@type":114},"It extends the Shefﬁeld Wargames Corpus from Interspeech 2013 with added fully annotated data (including female speech to improve gender bias) and includes blog-based language model data plus a Kaldi baseline system.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},517119,1791406731,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":24,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145,"preview_page_count":19,"preview_urls":146},24464137899374,"https://us-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2","INTERSPEECH 2016  \nSeptember 8–12, 2016, San Francisco, USA  \nThe Shefﬁeld Wargame Corpus-Day Two and Day Three  \nYulan Liu 1y, Charles Fox2, Madina Hasan 1, Thomas Hain 1yy  \n1MINI, SpandH, The University of Shefﬁeld, UK  \n2The University of Leeds, UK  \n[y](yacp12yl@sheffield.ac.uk)[acp12yl@sheffield.ac.uk](yacp12yl@sheffield.ac.uk) , [yy](yy t.hain@sheffield.ac.uk)[ t.hain@sheffield.ac.uk](yy t.hain@sheffield.ac.uk)  \nAbstract  \nImproving the performance of distant speech recognition is of considerable current interest, driven by a desire to bring speech recognition into people's homes. Standard approaches to this task aim to enhance the signal prior to recognition, typically using beamforming techniques on multiple channels. Only few real-world recordings are available that allow experimentation with such techniques. This has become even more pertinent with recent works with deep neural networks aiming to learn beamforming from data. Such approaches require large multichannel training sets, ideally with location annotation for moving speakers, which is scarce in existing corpora. This paper presents a freely available and new extended corpus of English speech recordings in a natural setting, with moving speakers. The data is recorded with diverse microphone arrays, and uniquely, with ground truth location tracking. It extends the 8.0 hour Shefﬁeld Wargames Corpus released in Interspeech 2013, with a further 16.6 hours of fully annotated data, including 6.1 hours of female speech to improve gender bias. Additional blog-based language model data is provided alongside, as well as a Kaldi baseline system. Results are reported with a standard Kaldi conﬁguration, and a baseline meeting recognition system. Index Terms: distant speech recognition, multi-channel speech recognition, natural speech corpora, deep neural network.  \n1. Introduction  \nMulti-channel based speech enhancement has been shown to be effective for Distant Speech Recongition (DSR), in both classical HMM-GMM systems and state-of-art Deep Neural Networks (DNNs) based systems. Compared to using recordings from single distant microphone only, beamforming is reported to reduce word error rate (WER) by 6-10% relative in large vocabulary conversational speech recognition tasks [1–3], and up to 60% relative in speciﬁc tasks [4, 5] . Multi-channel dereverberation brings an extra 20% relative WER reduction over single channel dereverberation [6] . Recently progress in neural networks have introduced further performance improvement in a variety of tasks, particularly from three aspects: progress in novel network structures [7, 8], application-oriented neural network structure and parameter manipulation [9–12], and data manipulation for neural network training [1, 13] . While the overall WERs keep going down, the recognition performance gap remains between using recordings from close-talking microphones and from distant microphones. To reduce this gap, research effort has focused on three approaches: developing novel structures to better utilize multichannel recordings in DNN [14, 15], employing task dependent meta information [16, 17], and simulating training data for speciﬁc DSR tasks [6, 18] . However research progress is limited by lack of data that provides multichannel distant recordings accom-  \npanied with headset recordings and speaker location tracking, in a natural speech setting where speakers are allowed to move freely. To address this problem, the present study extends the ﬁrst Shefﬁeld Wargames Corpus (SWC1,[19]) with more natural speech recordings from both headsets and distant microphones in moving talker conditions, accompanied with real time speaker location tracking.  \nThe paper is organized as follows. §2 reviews related work,§3 provides basic information about the set-up for the new recording days SWC2 and SWC3 . §4 discusses dataset deﬁnitions for two different ASR tasks: adaptation and standalone training. The details about language models (LMs) are i","cbCaipEzp39U8iXm","https://ap.wps.com/l/cbCaipEzp39U8iXm","pdf",4581673,"English","# Abstract\n# 1. Introduction\n# 2. Multi-channel Recordings in DSR","[{\"question\":\"What problem does the paper address in distant speech recognition?\",\"answer\":\"The paper targets the performance gap between close-talking microphones and distant microphones, where existing approaches and especially deep-learning beamforming methods are limited by scarce real-world multichannel data with location information.\"},{\"question\":\"What is new about the corpus introduced in this work?\",\"answer\":\"It provides a freely available extended English corpus recorded in natural settings with moving speakers, diverse microphone arrays, and ground-truth speaker location tracking.\"},{\"question\":\"How is the dataset organized and what additional resources are provided?\",\"answer\":\"It extends the Shefﬁeld Wargames Corpus from Interspeech 2013 with added fully annotated data (including female speech to improve gender bias) and includes blog-based language model data plus a Kaldi baseline system.\"}]","The Shefﬁeld Wargame Corpus-Day Two and Day Three - Interspeech 2016 Distant Speech Recognition Corpus | PDF",1791335602,13,[147,151,155,159],{"page_no":8,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/8a56ef34bbc0c8d397cfb3744560a586.webp",983,1390,{"page_no":14,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/116445191b45629b0465767ef17944f6.webp",982,1388,{"page_no":81,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/e3546d172e5d6c25ee4081eab96d4baf.webp",981,1387,{"page_no":19,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/54d75e6228c075388f17392d08011b90.webp",980,1385]