ContentsFrom Speech Recognition Towards Speech Understanding Outline Signal-Symbol-Signal Transformations in Spoken Dialog Systems Three Levels of Language Processing Challenges for Language Engineering Telephone-based Dialog Translation Context-Sensitive Speech-to-Speech Translation Dialog Translation 1 Dialog Translation 2 Dialog Translation 3 Verbmobil II: Three Domains of Discourse Data Collection with Mulitiple Input Devices Extracting Statistical Properties from Large Corpora Verbmobil Partner The Control Panel of Verbmobil From a Multi-Agent Architecture to a Multi-Blackboard Architecture Multi-Blackboard/Multi-Agent Architecture A Multi-Blackboard Architecture for the Combination of Results from Deep and Shallow Processing Modules Integrating Shallow and Deep Analysis Components in a Multi-Blackboard Architecture VHG: A Packed Chart Representation of Partial Semantic Representations Robust Dialog Semantics: Deep Processing of Shallow Structures Semantic Correction of Recognition Errors Robust Dialog Semantics: Combining and Completing Partial Representations The Understanding of Spontaneous Speech Repairs Automatic Understanding and Correction of Speech Repairs in Spontaneous Telephone Dialogs Integrating a Deep HPSG-based Analysis with Probabilistic Dialog Act Recognition for Semantic Transfer The Dialog Act Hierarchy used for Planning, Prediction, Translation and Generation Combining Statistical and Symbolic Processing for Dialog Processing Statistical Dialog Act Recognition Learning of Probabilistic Plan Operators from Annotated Corpora Automatic Generation of Multilingual Protocols of Telephone Conversations Automatic Generation of Minutes Dialog Protocol Spoken Clarification Dialogs between the User and the Verbmobil System Competing Strategies for Robust Speech Translation Architecture of the Semantic Transfer Module Extensions of Discourse Representation Theory Three English Translations of the German Word "Termin" Found in the Verbmobil Corpus Entries in the Transfer Lexicon: German to English (Simplified) Context-Sensitive Translation Exploiting a Discourse Model The Use of Underspecified Representations The Control Panel of Verbmobil Integrating Deep and Shallow Processing: Combining Results from Concurrent Translation Threads A Context-Free Approach to the Selection of the Best Translation Result Learning the Normalizing Factors Alpha and Beta from an Annotated Corpus Example of a Linear Inequation Used for Offline Learning The Context-Sensitive Selection of the Best Translation Verbmobil: Long-Term, Large-Scale Funding and Its Impact SmartKom: Intuitive Multimodal Interaction The Architecture of the SmartKom Agent (cf. Maybury/Wahlster 1998) SmartKom: A Transportable and Transmutable Interface Agent SmartKom-Public: A Multimodal Communication Booth SmartKom-Mobile: A Handheld Communication Assistant SmartKom-Home/Office: A Versatile Agent-based Interface Speech-based Interaction with an Organizer on a WAP Phone (Voice In - WML out) Conclusion Conclusion Conclusion |
Author: Prof. Wolfgang Wahlster
E-Mail: wahlster@dfki.de |