Digital Assistant Intent Routing for Battery and Memory Conservation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital assistants face challenges in efficiently handling user intents, leading to increased battery consumption and memory usage on local devices, particularly when dealing with knowledge-seeking requests that require external processing.
Innovation Solution
A system that determines the intent type and delegates tasks to either a local digital assistant or a remote model based on intent type, minimizing local processing for knowledge-seeking intents and leveraging a remote model for efficient task execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If all user intents are processed by the local digital assistant, then the device can handle any type of request, but battery consumption and memory usage increase significantly
Solution Approach 1:
The system segments intent processing into two distinct paths: local digital assistant handling for routine intents and remote model processing for knowledge-seeking intents. This segmentation allows the device to maintain versatility while reducing local processing overhead by offloading specific intent types to remote servers.
Solution Approach 2:
The system introduces an intermediary classification mechanism that determines whether to process intents locally or remotely. This mediator evaluates each intent type and routes it appropriately, enabling the local device to maintain adaptability while consuming less energy by using the remote model as an intermediary for complex knowledge-seeking tasks.
2Loss of time
If knowledge-seeking intents are processed locally, then response time may be faster, but device memory usage increases
Solution Approach 1:
The system extracts knowledge-seeking intent processing from the local device and transfers it to remote servers. This extraction eliminates the need to maintain large knowledge bases and complex processing capabilities on the local device, significantly reducing memory usage while still providing comprehensive knowledge-seeking capabilities through remote processing.
3Device complexity
If the local digital assistant handles all intents, then no external processing is needed, but the device complexity increases
Solution Approach 1:
The system implements a hybrid architecture where the local digital assistant handles routine intents and the remote model handles knowledge-seeking intents. This multi-functional approach allows the overall system to maintain high adaptability and versatility while keeping the local device relatively simple by leveraging the remote model's capabilities for complex tasks.
Data Source
AI summary
Systems and processes for delegating tasks to electronic devices based on intents and associated applications are provided. For example, receiving an input from a user at an electronic device, wherein the input is associated with an intent and determining, based on the intent, an intent type. In accordance with a determination that the intent type is a first type, causing a digital assistant to perform an action associated with the intent, and in accordance with a determination that the intent type is a second type, different than the first type, transmitting instructions to a model remote to the electronic device, wherein the instructions cause the model to perform an action associated with the intent.