Virtual Personal Assistants (VPAs) such as Amazon Alexa, Google Assistant, and Apple’s Siri are essential in modern smart homes due to their ease of use and ability to perform a variety of tasks. Some challenges of VPA use include 1) difficulty for deaf or hard-of-hearing users who cannot rely on auditory commands; 2) difficulty in understanding speech patterns that VPAs do not recognize; and 3) safety risks associated with children potentially accessing inappropriate content. This project will develop novel technologies and mechanisms to enhance the usability and safety of human-VPA interaction in smart home environments. This research also supports broader social goals such as engaging STEM students in the development of advanced algorithms, improving computer science curricula at participating institutions, and strengthening awareness of accessibility and safety issues in technologies. This research project proposes a suite of innovations to improve usability and safety in human-VPA interactions. First, this project develops an end-to-end system that enables users to interact with VPAs using sign language as input, effectively bypassing voice-based interaction barriers. Second, this project proposes a novel acoustic testing tool to identify usability issues caused by VPA’s that are not trained on certain speech patterns. Finally, this project develops a situational multi-user access control mechanism that allows primary users (e.g., parents or homeowners) to specify acc