《数字十年和自然用户界面》由会员分享，可在线阅读，更多相关《数字十年和自然用户界面（24页珍藏版）》请在金锄头文库上搜索。
1、User Interface in the Digital DecadeKai-Fu LeeCorporate Vice PresidentMicrosoft CorporationGraphicalGraphicalUser InterfaceUser InterfaceUser Interface EvolutionCommand lineCommand line19851985PCPC19901990GUIGUIMultiple WindowsMultiple WindowsMenusMenus19951995InternetInternetHyperlinksHyperlinksSea
2、rch EnginesSearch EnginesGraphical User InterfaceWhat you see is what you getGUI GUI InteractionInteractionl lLow learning curve.Low learning curve.l lGreat for frequent actions.Great for frequent actions.l lDifficult for rich domainsDifficult for rich domainsTechnology Technology AdvancesAdvancesl
3、lBitmap display.Bitmap display.l lMouse.Mouse.l lHTTP / HTML.HTTP / HTML.EnablesEnablesGraphicalGraphicalUser InterfaceUser InterfaceNatural Natural User InterfaceUser InterfaceUser Interface EvolutionCommand lineCommand line19851985PCPC19901990GUIGUIMultiple WindowsMultiple WindowsMenusMenus1995199
4、5InternetInternetHyperlinksHyperlinksSearch EnginesSearch EnginesDigital DecadeDigital DecadeXMLXMLWeb ServicesWeb ServicesSmart devicesSmart devicesNatural LanguageNatural LanguageMultimodal (speech, ink)Multimodal (speech, ink)Personal AssistantPersonal AssistantNatural User InterfaceDo what I mea
5、nNUI NUI InteractionInteractionl lNatural.Natural.l lScalable.Scalable.l lExpressive.Expressive.Technology Technology AdvancesAdvancesl lSmart devices.Smart devices.l lXML.XML.l lWeb services.Web services.EnablesEnables“A transforming capability that allows you to speak “A transforming capability th
6、at allows you to speak to your computer and it will understand what youre to your computer and it will understand what youre saying in context.”saying in context.”- Gordon Moore, 2002- Gordon Moore, 2002Super-Moores LawEnable human-level speech recognition (SR):Enable human-level speech recognition
7、(SR): Leveraging Moores Law + more data + research.Leveraging Moores Law + more data + research. Predictable 10% error reduction or more per year.Predictable 10% error reduction or more per year. Human-level performance possible in 10-20 years.Human-level performance possible in 10-20 years.Also ena
8、ble natural language, Bayesian learning.Also enable natural language, Bayesian learning.TasksSR error rateHuman error rateSR-Human GapFree style transcription30%4%19 yearsAlphabet letters5%1%15 yearsRead newspaper transcription3%0.9%11 yearsMicrosofts Practical NUI PlanStart with natural entry point
9、s:Speech for telephony.Speech for telephony.Typing for search and help.Typing for search and help.Provide simple, evolutionary UI for end-users.Users will perceive NUI as “better search”.Users will perceive NUI as “better search”.Search/help needs to complement GUI.Search/help needs to complement GU
10、I.Build on .NET:Visual Studio for development.Visual Studio for development.NET frameworks, XML, web services.NET frameworks, XML, web services.A Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation &
11、 Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalA Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-M
12、odalBuy 100 shares of Microsoft.A Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalBuy Microsoft stock.HighHighInternetIn
13、ternetTVTVPhonePhonePDAPDAEase of text input (keyboard/pen)Ease of text input (keyboard/pen)Ease Ease of GUIof GUI(screen/(screen/Pointer)Pointer)LowLowHighHighPCPCTabletTabletPCPCScreenScreenPhonePhoneScreenScreenPhonePhonePDAPDATabletTabletPCPCCarCarCarCarInternetInternetTVTVSmart Devices Needs NU
14、I (speech)PhonePhonePCPCScreenScreenPhonePhonePDAPDATabletTabletPCPCCarCarInternetInternetTVTVA Roadmap for SpeechEase of text input (keyboard/pen)Ease of text input (keyboard/pen)Ease Ease of GUIof GUI(screen/(screen/Pointer)Pointer)HighHighHighHighLowLowSpeech-Only TelephonyDictationMultimodal Com
15、mand/ControlSpeech Demonstrations.NET Speech Platform for TelephonySupport telephony speech from web application.Support telephony speech from web application.Reduces cost of phone call from $2 to $0.10.Reduces cost of phone call from $2 to $0.10.Uses .NET Framework and Visual Studio.Uses .NET Frame
16、work and Visual Studio.Supports multimodal UI using SALT.Supports multimodal UI using SALT.Beta in June; RC in October.Beta in June; RC in October.Tablet PC SpeechSpeech dictations is faster than handwriting.Speech dictations is faster than handwriting.Supports multimodal UI.Supports multimodal UI.A
17、 Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalFind email from John about the budget.A Roadmap Towards NUI VisionTTS S
18、R DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalMy printer is stuck.A Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help T
19、asks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalPrint 10,000 copies in Kinkos Beijing.A Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&ATex
20、t onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalWhat time does Bill Gates talk end?A Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelega
21、tion & Pro-ActiveFederation Agents PlanningFederation Agents PlanningMulti-ModalSend flowers to my wife on her birthdayA Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Age
22、nts PlanningFederation Agents PlanningMulti-ModalHold all calls unless its urgent.A Roadmap Towards NUI VisionTTS SR DialogTTS SR DialogTelephony speechSearch Help Tasks Q&ASearch Help Tasks Q&AText onlyDelegation & Pro-ActiveDelegation & Pro-ActiveFederation Agents PlanningFederation Agents Plannin
23、gMulti-ModalI want to plan a vacation to Europe.Structured Storage needs NUIFind email from John Find email from John about the Budgetabout the BudgetSELECT DocumentNameSELECT DocumentNameFROM Local, LocalMailSystem, LocalDocumentsFROM Local, LocalMailSystem, LocalDocumentsWHERE Author contains “Joh
24、n” and Subject contains WHERE Author contains “John” and Subject contains LSP_Expand(“Budget”)LSP_Expand(“Budget”)ORDER BY ModifiedDate, CreateDateORDER BY ModifiedDate, CreateDateMy printer is stuckMy printer is stuckSELECT DocumentNameSELECT DocumentNameFROMLocal, LocalHardware, HardwareVendorSupp
25、ortFROMLocal, LocalHardware, HardwareVendorSupportWHERE Vendor=HP and Model=DeskJet 550” and Body contains WHERE Vendor=HP and Model=DeskJet 550” and Body contains LSP_Expand(“jam”)LSP_Expand(“jam”)ORDERED BY Context, LearnedBehavior.ORDERED BY Context, LearnedBehavior.Show me the new Show me the ne
26、w Dell notebookDell notebookSELECT DocumentNameSELECT DocumentNameFROM WebSearch, UserHeuristicsFROM WebSearch, UserHeuristicsWHERE Vendor=Dell and Category=laptops and Age60WHERE Vendor=Dell and Category=laptops and Age60Make the letters on Make the letters on my screen biggermy screen biggerSELECT
27、 TaskNamesSELECT TaskNamesFROM LocalTasks, OSTasks, USerHeuristicsFROM LocalTasks, OSTasks, USerHeuristicsWHERE Name contains Video ResolutionWHERE Name contains Video ResolutionORDER by Context, LearnedBehaviorORDER by Context, LearnedBehaviorPlay BritneyPlay BritneySELECT MediaNameSELECT MediaName
28、FROM Local, WebMediaProvidersFROM Local, WebMediaProvidersWHERE Name contains Britney OR Name contains SpearsWHERE Name contains Britney OR Name contains SpearsORDER BY CurrentDirectoryBias, LearnedBehavior, Name, ContextORDER BY CurrentDirectoryBias, LearnedBehavior, Name, ContextNatural language d
29、emonstrationsSmart search & help.Users dont know where to go or what to type.Users dont know where to go or what to type.Companies dont know whats on users minds.Companies dont know whats on users minds.We want to close that gap.We want to close that gap.Q&A for “any question”.Unstructured Unstructu
30、red structured tool is needed. structured tool is needed.Combines statistical and natural language.Combines statistical and natural language.ConclusionNatural UI is needed for the Digital Decade:Scalable, natural UI for smart devices.Scalable, natural UI for smart devices.Retrieval of information fr
31、om structured storage.Retrieval of information from structured storage.UI to connected web services.UI to connected web services.Natural UI will arrive as an evolution:Telephony speech, search, help are the first applications.But in 10 years, Natural UI will be viewed as the largest revolution since Graphical UI. 2001 Microsoft Corporation. All rights reserved. 2001 Microsoft Corporation. All rights reserved.
