Intelligent Assistant Proactive Contextual Actions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intelligent automated assistants face challenges in processing unrestricted natural-language user inputs, which can be slow, processor-intensive, and power-intensive, and users may avoid using voice commands due to cognitive or social burdens, leading to manual access of device functionality.
Innovation Solution
The system provides proactive surfacing and performance of contextual actions by an intelligent automated assistant, using current contextual information, such as visual context, to identify and initiate relevant actions without explicit user requests, reducing the need for natural-language inputs and minimizing processing power and cognitive burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the digital assistant processes unrestricted natural-language user inputs to determine and respond to user intents, then the assistant can provide comprehensive functionality access, but the processing becomes slow, processor-intensive, and power-intensive
Solution Approach 1:
The system proactively identifies and surfaces contextual actions before the user explicitly requests them. By analyzing contextual information (visual, auditory, sensor data) in advance, the assistant prepares relevant actions and presents them to the user, eliminating the need for the user to formulate detailed natural-language requests and reducing real-time processing requirements
Solution Approach 2:
The digital assistant autonomously monitors contextual information and automatically identifies relevant actions without requiring continuous user input. The system serves itself by maintaining awareness of the environment and user context, proactively generating action suggestions, and managing the interaction flow, thereby reducing the computational burden of processing unrestricted natural-language inputs
2Ease of operation
If the user provides natural-language requests to the digital assistant, then the assistant can understand user intent, but the user experiences cognitive and social burden
Solution Approach 1:
The system introduces contextual information (visual data from cameras, sensor readings, device state) as an intermediary between the user and the digital assistant. Instead of requiring the user to articulate their intent through natural language, the assistant observes the environment and infers user needs from contextual cues, acting as a mediator that translates environmental context into actionable insights
Solution Approach 2:
The digital assistant autonomously gathers and analyzes contextual information from multiple sources (camera feeds, sensors, device state) without requiring user input. The system self-services by continuously monitoring the environment, interpreting contextual data, and generating relevant action suggestions, thereby eliminating the cognitive burden of formulating natural-language requests
3Use of energy by moving object
If the digital assistant proactively identifies and performs contextual actions based on visual context information, then the assistant reduces processing power consumption and user cognitive burden, but the system must continuously monitor and interpret contextual data
Solution Approach 1:
The system extracts only the most relevant contextual information from the environment rather than processing all available data continuously. By selectively identifying and focusing on key contextual cues (such as specific visual elements or sensor readings that indicate user intent), the assistant reduces the overall processing load while maintaining effective proactive action identification
Solution Approach 2:
The digital assistant applies different levels of processing intensity to different types of contextual information based on their relevance. High-priority contextual data (such as direct visual indicators of user intent) receive intensive analysis, while lower-priority data receive minimal processing. This localized quality approach optimizes energy consumption by concentrating computational resources where they are most needed
Data Source
AI summary
Systems and processes for operating an intelligent automated assistant are provided. In some embodiments, after providing an initial output with information on a particular topic, an output including updated information on the topic is provided in response to an input interacting with a digital assistant. In some embodiments, based on contextual data including visual context data, an action is identified and proactively performed.