Skill Enablement via Trust Categorization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speech recognition systems face challenges in automatically enabling relevant skills while preventing fraudulent or irrelevant skills from being activated, particularly when user inputs are ambiguous or transcription errors occur, leading to potential malicious activities or incorrect skill invocation.
Innovation Solution
The system categorizes skills based on user inputs, device usage, and published events, allowing for automatic enablement of trusted skills and blocking untrusted ones, using a categorization process that involves user feedback, machine learning models, and event monitoring to determine skill trust levels and discoverability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If automatic skill enablement is implemented based on speech recognition, then user convenience and interaction speed are improved, but the risk of fraudulent or incorrect skill invocation increases
Solution Approach 1:
The patent introduces skill categorization as an intermediary layer between speech recognition and skill invocation. A categorization component classifies skills into different categories (e.g., trusted, untrusted, ambiguous) based on multiple factors including speech recognition confidence, user feedback, and device usage patterns. This intermediary classification system enables automatic enablement for trusted skills while preventing fraudulent or ambiguous skill invocation, thus resolving the contradiction between convenience and reliability.
2Measurement precision
If skill categorization uses multiple factors including user feedback and event monitoring, then skill trust determination accuracy is improved, but system complexity increases
Solution Approach 1:
The patent segments the skill categorization process into distinct functional components: a categorization component that classifies skills, a feedback component that collects user feedback, an event monitoring component that tracks device usage events, and a machine learning component that processes data. Each component performs a specific function and can be independently optimized or modified. This segmentation allows the system to achieve high measurement precision through multiple factors while managing complexity through modular design.
3Manufacturing precision
If machine learning models are used to analyze user feedback and device usage, then skill categorization accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The patent implements preliminary action by pre-processing and storing user feedback and device usage events as they occur, rather than analyzing them in real-time when skill invocation is needed. The machine learning component trains models on accumulated historical data to create pre-computed skill categories and trust levels. This allows the system to maintain high categorization precision while reducing real-time processing time, as the heavy computational work is performed in advance.
Data Source
AI summary
Techniques for selecting a skill, to respond to a user input, using skill rankings are described. A skill's ranking may be determined in different manners. In one example, a skill's ranking may be determined based on a number of different users inputting commands that invoke the skill over a period of time. In another example, a skill's ranking may be determined based on a number of different devices that capture user inputs that invoke the skill over a period of time. A system may determine whether to automatically enable a skill (e.g., without user input received after the original user input), or ask the user whether the skill should be enabled, based on the skill's ranking. Moreover, a system may use a skill's ranking to determine whether to interact with a user, to enable the skill, using a graphical user interface or a voice user interface.