This research introduces a novel system that integrates spatial speech understanding into large language models to create contextually aware applications for wearable technologies.

• Leverages microstructure-based spatial sensing to extract precise Direction of Arrival (DoA) information using a single microphone
• Combines spatial audio processing with LLM capabilities to enable more natural and context-aware interactions
• Creates new possibilities for adaptive applications in wearable devices that understand both what users say and where speech is coming from
• Addresses engineering challenges in real-time spatial processing on resource-constrained wearable devices

This innovation represents a significant advancement for engineering wearable technology with potential to transform how we interact with AI assistants in our daily lives.

Spatial Audio Processing with Large Language Model on Wearable Devices