设计方法和实现审的部分关键技术。给出了系统实现的鼷桀和模垄,弗对系统送行了分析和评价。(6)将分毒筑语音识别技术和WebServices技本络会,挺遗了努蠢式语音识剿服务静概念,并输出了实臻模蹩,获SOAP、WSDL鞍UDDI三方面蜜瑷了分布式谮音识别服务的框架。关键{蘑:语音测麓器,分布斌语音识别。绷挤控镑l,熬错控制,语费可扩展标识谮言,;WithWebServicestechnology,·technology,essIntemetonlywithkeyboards,essInteractthroughvoice,,badlystabilityofrecognitionresults,work,SOsince1999tillnow,theapplicationsofVoiceBrowserareonlyinthespecialfieldsthatspeechrecognitionraterelativehi曲,suchasstockquery,weatherforecastandscheduledflightquery,itcan’,theDistributedSpeechRecognitiontechnologyhasattractedalotofresearcher’sattention,municationchannel,municationspeechrecognitionratehighandsupportspeoplewithsteadilyrecognitionresults。,theDistributedSpeechIkcognitiontechnologyisthoroughlysurveyed,aseriesofworksaredevelopedonthefollowingtopics:thetheoriesandmethodsaboutDistributedSpeechRecognition,thewaystosupportthedistributedVoiceBrowserbasedonVoiceXML,andtheDistributedSpeechRecognitionservices。Thewholeworksaremainlyconcludedinthefollowingaspects:Firstly,,,ingsoflowtelephonespeechrecognitionrate,work,expandstheapplicationofVoiceBrowsertotheequipmentbasedonvoice,impliesthe‘'work'’.Secondly,basedonthestudyofthecorrelationsofneighborflames,,,workstatus,,bytakingadvantageofRSencoding,—channelencodingscheme
基于分布式语音识别地语音浏览器研究 来自淘豆网m.daumloan.com转载请标明出处.
