Localized Voice Assistant for Meeting Device Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice assistants for operating devices in meeting settings require cloud-based processing and internet connectivity, leading to latency and risks of data leakage due to persistent voice data retention, and are often unavailable when internet connections are poor.
Innovation Solution
A localized voice assistant system that processes voice commands locally, detecting devices at a location, associating them with commands, and performing operations without internet or cloud-based processing, reducing latency and data retention risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If cloud-based processing is used for voice commands, then processing power and intelligence are improved, but latency increases and data security deteriorates
Solution Approach 1:
The system segments voice processing into two parts: local wake word detection and command recognition are performed on the device itself, while more complex tasks can optionally use cloud resources. This segmentation allows immediate local responses while maintaining option for enhanced processing when needed.
Solution Approach 2:
The patent introduces an on-device voice processing intermediary that sits between the microphone and the cloud connection. This intermediary can process and filter voice data locally, determining what needs to be sent to the cloud, thereby reducing unnecessary data transmission and enabling faster local responses.
2Extent of automation
If cloud-based processing is used for voice commands, then processing capability is improved, but data security and privacy deteriorate due to persistent data retention
Solution Approach 1:
The patent extracts the essential voice processing functionality from the cloud and places it directly on the device. By taking out the critical processing tasks from cloud-based systems, the invention eliminates the need for persistent data retention and transmission, thereby removing the data leakage risk while maintaining processing capability.
Solution Approach 2:
The system uses ephemeral, in-memory processing for voice data that is immediately discarded after local processing. This approach treats voice data as temporary rather than persistent, eliminating the security risks associated with long-term data retention while still enabling effective voice command processing.
3Adaptability or versatility
If internet connectivity is required for voice assistant operation, then access to external services is improved, but system availability deteriorates when connection is poor or lost
Solution Approach 1:
The system performs preliminary action by pre-loading wake word models and basic command recognition capabilities directly onto the device before operation begins. This preliminary local processing capability ensures the system remains available and responsive even when internet connectivity is unavailable, while still providing pathways to external services when connected.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
An embodiment sets forth a technique for controlling a device at a location via a localized assistant application that processes locally inputs for controlling the device. The technique includes detecting a device at a location, associating the device with at least one device command, receiving an input, processing the input locally to determine a device command associated with the input and included in the at least one device command, and causing at least one operation to be performed by the device in accordance with the device command..