Intelligent Assistant Language Understanding On Device
This addresses the need for efficient and private on-device AI assistants for users, though it appears incremental as it builds on existing technologies without introducing a new paradigm.
The paper tackles the problem of running natural language understanding for personal digital assistants directly on devices, resulting in a system that is more private, reliable, faster, expressive, and accurate compared to server-based alternatives.
It has recently become feasible to run personal digital assistants on phones and other personal devices. In this paper we describe a design for a natural language understanding system that runs on device. In comparison to a server-based assistant, this system is more private, more reliable, faster, more expressive, and more accurate. We describe what led to key choices about architecture and technologies. For example, some approaches in the dialog systems literature are difficult to maintain over time in a deployment setting. We hope that sharing learnings from our practical experiences may help inform future work in the research community.