How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U

How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U

ID:38267638

大小:107.15 KB

页数:4页

时间:2019-05-27

How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U_第1页
How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U_第2页
How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U_第3页
How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U_第4页
资源描述:

《How NLP techniques can improve speech understanding ROMUS – a Robust Chunk based Message U》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库

1、HowNLPtechniquescanimprovespeechunderstanding:ROMUS–aRobustChunkbasedMessageUnderstandingSystemUsingLinkGrammarsJérômeGoulian,Jean-YvesAntoine,FranckPoirierVALORIALaboratory,UniversityofSouth-BrittanyYvesCoppensResearchCenter,TohannicF-56000VANNES,Francejerome.goul

2、ian@univ-ubs.frAbstractquestionmaybe,thispaperinvestigateswhetherlinguisticallysophisticationintheanalysiscanimprovespeechunderstand-Thispaperdiscussestheissueofhowaspeechunderstandingingwhilepreservingrobustness.Weargueitispossibletoob-systemcanbemaderobustagainst

3、spontaneousspeechphe-tainmoredetailedrepresentationsthankstoacombinationofanomena(hesitationsandrepairs)aswellasachievingadetailedsyntactically-followedbyasemantically-drivenrobustparsinganalysisofspokenFrench.TheRomussystemispresented.Itwithareasonableamountofeffo

4、rt.Wewillshowwecancap-implementsspeechunderstandinginatwo-stageprocess.TheturethereforeamajorityofinformationthatcanbeemployedforÞrststageachievesaÞnite-stateshallowparsingthatconsistsdisambiguation,detectionandprocessofrepairs,andthatcaninsegmentingtherecognizedse

5、ntenceintobasicunits(spoken-beusefulforcontextinterpretation.adaptedchunks).Thesecondone,aLinkGrammarparser,looksItwasdemonstratedthatparsingnaturallanguagecanbeforinter-chunksdependenciesinordertobuildarichrepresen-handledbyincrementalapproachesbasedforinstanceont

6、hetationofthesemanticstructureoftheutterance.Thesedepen-principleofÒchunkingÓtheinputintosmallandeasilymanage-denciesaremainlyinvestigatedatapragmaticlevelthroughtheableunits[6].ROMUS,thespeechunderstandingsystemweconsiderationofataskconcepthierarchy.Discussionabou

7、tthepresentinthispaper,appliesandsensiblyadaptstheseideastoapproachadopted,itsbeneÞtsandlimitations,isbasedonthespontaneousspokendialogue.ROMUSachievesspeechunder-resultsofthesystemÕsassessmentcarriedoutunderdifferentstandinginatwostageprocess.TheÞrststagesegmentst

8、helinguisticphenomenaduringanevaluationcampaignheldbyutteranceingrammaticalunits.TheseunitsarethenattachedtheFrenchCNRS.togetherusingaLinkGrammar

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。