ABSTRACT
Designing a SUI involves two levels of design decisions. First, you need to make certain high-level design decisions regarding system-level interface properties. Only then can you get down to the details of designing speci‹c system prompts and dialogs. The high-level decisions include:
• Choosing the barge-in style • Selecting recorded prompts or synthesized speech • Deciding whether to use audio formatting • Using simple or complex speech recognition • Adopting a concise or verbose prompt style • Allowing only speech input or speech plus touchtone • Choosing a set of global navigation commands • Deciding when and how to transfer to human agents • Choosing a help mode or self-revealing help
There often is no single correct answer; the appropriate decisions depend on the application, the users, and the users’ environment(s). The remainder of this chapter presents the trade-offs associated with each of these decisions.
