TAILIEUCHUNG - Báo cáo khoa học: "MIMUS: A Multimodal and Multilingual Dialogue System for the Home Domain"

This paper describes MIMUS, a multimodal and multilingual dialogue system for the in– home scenario, which allows users to control some home devices by voice and/or clicks. Its design relies on Wizard of Oz experiments and is targeted at disabled users. MIMUS follows the Information State Update approach to dialogue management, and supports English, German and Spanish, with the possibility of changing language on–the– fly. MIMUS includes a gestures–enabled talking head which endows the system with a human–like personality. . | MIMUS A Multimodal and Multilingual Dialogue System for the Home Domain J. Gabriel Amores Julietta Research Group Universidad de Sevilla jgabriel@ Guillermo Perez Julietta Research Group Universidad de Sevilla gperez@ Pilar Manchon Julietta Research Group Universidad de Sevilla pmanchon@ Abstract This paper describes MIMUS a multimodal and multilingual dialogue system for the inhome scenario which allows users to control some home devices by voice and or clicks. Its design relies on Wizard of Oz experiments and is targeted at disabled users. MIMUS follows the Information State Update approach to dialogue management and supports English German and Spanish with the possibility of changing language on-the-fly. MIMUS includes a gestures-enabled talking head which endows the system with a human-like personality. 1 Introduction Figure 1 MIMUS Architecture This paper describes MIMUS a multimodal and multilingual dialogue system for the in-home scenario which allows users to control some home devices by voice and or clicks. The architecture of MIMUS was first described in Perez et al. 2006c . This work updates the description and includes a life demo. MIMUS follows the Information State Update approach to dialogue management and has been developed under the EU-funded TALK project Talk Project 2004 . Its architecture consists of a set of OAA agents Cheyer and Martin 1972 linked through a central Facilitator as shown in figure 1 The main agents in MIMUS are briefly described hereafter The system core is the Dialogue Manager which processes the information coming from the different input modality agents by means of a natural language understanding module and provides output in the appropriate modality. wrapper for Nuance. Currently the system supports English Spanish and German with the possibility of changing languages on-the-fly without affecting the dialogue history. The HomeSetup agent displays the house layout with all the devices and their state. Whenever

• Báo cáo khoa học
• A Multimodal
• Multilingual Dialogue System
• the Home Domain
• báo cáo khoa học
• mô hình ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Multimodal discourse analysis
• Multimodal discourse
• Discourse analysis
• Analysis the picture
• Interpersonal meaning
• Multimodal instruction
• Initial teacher training
• Prospects and challenges
• Multimodal instruction in initial teacher training
• Incorporating multimodal practices
• Multimodal user interfaces
• Experimental evaluation
• Pursuit tracking
• Human factors
• Multimodal feedback
• International Journal of computer science & communication networks
• Three level multimodal medical image fusion
• Medical image fusion
• Fuse lowand high frequency coefficients
• Multimodal medical imaging
• Attachment based intervention
• Multimodal pain therapy
• Chronic pain
• Multimodal treatment program
• Therapeutic outcome
• Multimodal discourse analysis of fast food advertisements in English
• Fast food advertisements in English
• Both theoretical and analytical framework
• Non verbal elements
• Fast food ads
• BMC Cancer
• Aerobic training
• Anthroposophic medicine
• Cognitive behavior therapy
• Breast cancer
• Cancer related fatigue
• Eurythmy therapy
• Multimodal intervention
• Multimodal transport
• Intermodal transport
• Combined transport
• Direct multimodal transport
• Inland water transport
• Container transport
• Romantic comedy movie posters
• Analyzing multimodal texts
• Quite rapid pace
• BMC Anesthesiology
• Transversus abdominis plane block
• Oral analgesics
• Caesarean deliver
• Postoperative pain relief
• Multimodal oral analgesia
• Multimodal Dialogue
• Domain Reasoning
• Dialogue Management
• Conclusion
• Wearable Computing
• Conversational Systems
• Classification of Feedback Expressions
• Multimodal Data
• Costanza Navarretta Patrizia Paggio
• báo cáo ngôn ngữ
• ngôn ngữ tự nhiên
• Improving On line Handwritten Recognition
• Translation Models
• Multimodal Interactive Machine Translation
• Towards a Framework
• Abstractive Summarization
• Multimodal Documents
• Multimodal Menu based Dialogue
• Speech Cursor in DICO II+
• Staffan Larsson
• Learning Effective Multimodal Dialogue Strategies
• Wizard of Oz data
• Bootstrapping and Evaluation
• Automated Vocabulary Acquisition
• Interpretation in Multimodal
• A Multimodal Interface
• Access to Content in the Home
• Michael Johnston
• Discriminating image senses
• clustering with multimodal features
• Nicolas Loeff
• Using Machine Learning
• Explore Human Multimodal
• Clarification Strategies
• Predicting Evidence of Understanding
• Monitoring User’s Task Manipulation
• Multimodal Conversations
• Bi Directional Parsing
• Generic Multimodal Interaction
• Frank Rudzicz
• A multimodal system
• multimedia meeting browsing and retrieval
• Agnes Lisowska