Interactive Multimodal Interfaces: Designing for Human Performance

*9530666 Oviatt这项研究为开发更强大、更灵活、更易访问、更以用户为中心的界面提供了基础，这些界面适用于结合了人类语言技术的多模式/多媒体系统。具体目标包括：(1)在错误解决过程中对用户适应的语言进行建模，并将这些信息应用于设计更好的错误避免和解决技术；(2)检查不同类型和质量的复杂视觉显示对用户向交互系统的语言输入的影响，并设计有效而透明地引导用户输入以匹配系统处理能力的界面；以及(3)开发可应用于良好集成的多通道界面设计的认知工作量的精细化度量。为了实现这些目标，正在使用一种快速、半自动的模拟技术，该技术支持在多模式人机交互期间收集高保真数据。这项研究将极大地推进多模式界面设计所需的经验、理论和应用基础，并将对建立能够在实际现场环境中取得成功的可用系统至关重要。特别是，它可以对提高广泛应用的下一代通信设备和移动技术的商业可行性产生重大影响。***

*** 9530666 Oviatt A foundation is provided by this research for the development of substantially more robust, flexible, accessible, and user-centered interfaces for multimodal/multimedia systems that incorporate human language technology. Specific objectives include: (1) modeling users' adapted language during error resolution, and applying this information to the design of better error avoidance and resolution techniques, (2) examining the impact of different types and qualities of complex visual display on users' linguistic input to interactive systems, and designing interfaces that effectively but transparently guide users' input to match system processing capabilities, and (3) developing a refined metric of cognitive workload that can be applied to the design of well-integrated multimodal interfaces. To pursue these objectives, a rapid, semi-automatic simulation technique is being used that supports the collection of high-fidelity data during multimodal human-computer interaction. This research will significantly advance the empirical, theoretical, and applied foundations needed for the design of multimodal interfaces, and will be critical to establishing usable systems that can succeed in actual field settings. In particular, it can have a major impact on improving the commercial viability of next-generation communications devices and mobile technology for a wide range of applications. ***