Ear-Wearable Local AI Assistant for Offline Voice Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice assistant systems require network connectivity to function, limiting their usability when a network connection is unavailable, and they lack the ability to handle queries locally without relying on network resources.
Innovation Solution
An ear-wearable device equipped with speech recognition and generation capabilities, along with a local database, allows for locally-handled queries and actions, and can connect to a gateway device for network assistance when needed, enabling independent operation and seamless integration with network resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If voice assistant systems rely on network connectivity to function, then they can access comprehensive resources and services, but they become unusable when network connection is unavailable
Solution Approach 1:
The system segments query handling into two distinct paths: local processing for simple queries using the input database and local resources, and network processing for complex queries requiring external resources. This segmentation allows the system to maintain functionality in both networked and offline modes, resolving the contradiction between versatility and reliability.
Solution Approach 2:
The system performs preliminary actions by pre-loading commonly needed information into the local input database and caching recent interactions. This preliminary preparation enables the device to handle routine queries locally without network connectivity, ensuring basic functionality remains available even when offline.
2Reliability
If the system processes all queries locally, then network connectivity is not required, but the system cannot access external resources or handle complex queries
Solution Approach 1:
The system dynamically adjusts its processing mode based on query complexity and network availability. Simple, routine queries are handled locally using the input database and local resources, while complex queries requiring external information are routed to network resources. This dynamic adaptation resolves the contradiction between local reliability and network-based versatility.
Solution Approach 2:
The gateway device serves as an intermediary between the ear-wearable device and network resources. It receives queries from the wearable device, determines whether to process them locally or forward them to network resources, and returns appropriate responses. This intermediary architecture enables seamless integration of local and network-based processing.
3Adaptability or versatility
If the system continuously checks for network connectivity, then it can switch between local and network modes, but this increases system complexity and energy consumption
Solution Approach 1:
Instead of continuous connectivity checking, the system employs periodic or event-driven connectivity assessments. The gateway device checks network availability at appropriate intervals or in response to specific user actions, rather than continuously monitoring. This periodic approach reduces computational overhead and energy consumption while maintaining the ability to switch between local and network modes when needed.
Data Source
AI summary
Embodiments herein relate to a local assistant system responding to voice input using an ear-wearable device. The system detects a wake-up signal and receives a first voice input communicating a first query content. The system includes speech recognition circuitry to determine the first query content, speech generation circuitry, and an input database of locally-handled user inputs. If the first audio input matches one of the locally-handled user inputs, then the system takes a local responsive action. If the first audio input does not match one of the locally-handled user inputs, then the system transmits at least a portion of the first query content over a wireless network to a network resource.