Challenges and Methods in Design of Domain-Specific Voice Assistants

摘要

Most of the currently existing voice assistants, like Alexa, Siri, Google Assistant, and Cortana, are generalists. They act as a unifying voice interface to a myriad of controls but rarely support domain-specific expert functionalities. There are efforts to provide more targeted assistant experiences and capabilities around specific areas of applications. In this paper, we discuss several challenges and opportunities in the design of domain-specific voice assistants. We outline a variety of methods to create and utilize an understanding of domain-specific user language and ideas to prototype and study the envisioned user experiences.