Hub Device Voice Model Storage for Low-Latency Multi-Device Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-device systems, users face inconvenience in selecting devices to provide services based on voice inputs due to limitations in memory capacity, processing speed, and network usage fees, and existing on-device models are insufficient for efficient device control.
Innovation Solution
A hub device that stores a voice assistant model to determine and control devices based on user intentions, utilizing a device determination model and function determination model, with the ability to update and manage these models from a voice assistant server, optimizing resource allocation and device selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a voice assistant service-providing server is used to control devices based on voice input, then the hub device can provide voice assistant services, but network usage fees increase and response speed decreases
Solution Approach 1:
The patent extracts the voice assistant model from the remote server and stores it locally in the hub device's internal memory. This allows the hub device to independently perform voice recognition and device control functions without continuously communicating with the external server, thereby eliminating network usage fees while maintaining full voice assistant service capability.
Solution Approach 2:
The patent introduces an on-device voice assistant model as an intermediary between the user's voice input and the device control system. This local model acts as a mediator that processes voice commands independently, eliminating the need for remote server communication and reducing both network costs and response time.
2Speed
If an on-device voice assistant model is stored in internal memory, then network usage fees are reduced and response speed increases, but the model may not be stored due to memory capacity limitations
Solution Approach 1:
The patent applies partial action by storing only the essential voice assistant model in the hub device's internal memory, while maintaining the ability to update and supplement this model. The system stores a compressed or optimized version of the model that fits within memory constraints, performing the most critical voice recognition functions locally.
Solution Approach 2:
The patent implements preliminary action by pre-storing the voice assistant model in internal memory before it is needed for operation. The model is downloaded and stored in advance during device initialization or off-peak times, ensuring that when voice commands are received, the model is already available in memory for immediate processing without requiring real-time network access.
3Speed
If the voice assistant model is stored in internal memory, then response speed increases, but processing speed may be insufficient for complex model operations
Solution Approach 1:
The patent segments the voice assistant system into two parts: a compact core model stored in internal memory for fast local processing of common commands, and a more comprehensive model on the remote server for handling complex or uncommon voice inputs. This segmentation allows the system to achieve fast response times for frequent operations while maintaining the capability to handle complex tasks when needed.
Solution Approach 2:
The patent implements a dynamic model loading strategy where the system adapts its processing approach based on the complexity of the voice input. For simple, frequent commands, the compact local model provides immediate response. For complex or unrecognized commands, the system dynamically transitions to server-based processing, optimizing the balance between response speed and processing capability.
Data Source
AI summary
A method, performed by a hub device, of storing a voice assistant model for controlling a device includes receiving information about a first device connected to the hub device, requesting a voice assistant server to update a device determination model stored in the hub device; receiving the updated device determination model from the voice assistant server and storing the received updated device determination model; requesting, from the voice assistant server, information about a function determination model corresponding to the first device; receiving the information about the function determination model corresponding to the first device from the voice assistant server, and determining whether to store the function determination model in the hub device based on the received information; and based on the function determination model being determined to be stored in the hub device, storing, in the hub device, the function determination model corresponding to the first device.