Smart Home Device Resolution for Ambiguous Voice Commands
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice command systems struggle to accurately identify and control smart home devices when user utterances do not match specific naming indicators or correspond to multiple devices, leading to ambiguity in device operation.
Innovation Solution
A remote system employs a target-inference component utilizing linguistics, device-activation, device-affinity, and location models to generate features for a classification model, which ranks and identifies the most likely smart home device to be operated based on user utterances, historical data, and device states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If voice command systems use simple naming indicators for smart home devices, then ease of operation is improved, but device resolution accuracy deteriorates when user utterances do not match specific naming indicators
Solution Approach 1:
The patent introduces an intermediary classification model that acts as a mediator between the user's voice command and the specific smart home device. This model processes the voice utterance through multiple features (linguistics, device-activation, device-affinity, location) and generates a ranked list of candidate devices, resolving the ambiguity when direct naming indicator matching fails.
Solution Approach 2:
The system changes the parameter of device identification from simple string matching to a multi-dimensional classification approach. By evaluating multiple features (linguistic similarity, historical activation patterns, device affinity scores, location context), the system transforms the resolution process from exact match to probabilistic ranking, improving accuracy without complicating user input.
2Measurement precision
If voice command systems implement multiple models for device resolution, then device resolution accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the device resolution process into distinct components: a linguistics model for text analysis, a device-activation model for historical pattern recognition, a device-affinity model for user preference analysis, and a location model for spatial context. Each model independently processes specific features and contributes to the final classification, making the complex system manageable through modular design.
Solution Approach 2:
The classification model serves multiple functions: it ranks candidate devices, identifies the most likely target device, and provides a unified interface for processing diverse voice commands. This multi-functional approach consolidates the complexity into a single versatile component rather than requiring separate specialized systems for each function.
3Measurement precision
If voice command systems provide detailed device identifiers, then device resolution accuracy is improved, but ease of operation deteriorates as users must remember specific naming indicators
Solution Approach 1:
The system enables self-service device identification by automatically analyzing the voice utterance and determining the most likely target device without requiring the user to provide precise device names. The classification model performs the identification work autonomously by evaluating multiple features and ranking candidate devices, freeing the user from the burden of remembering specific naming indicators.
Data Source
AI summary
Systems and methods for target-device resolution are disclosed. A user may speak a user utterance requesting an action to be performed with respect to an accessory device, such as a smart-home device. The user utterance may include an identifier for the accessory device, but that identifier may not correspond to a naming indicator of an accessory device and/or may correspond to multiple naming indicators. In these examples, one or more models are utilized to identify which accessory device is most likely to correspond to the accessory device targeted by the user utterance for operation.