AI Voice Response Timing Based on User Engagement

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Digital assistants lack the capability to determine appropriate times for executing voice commands, often delivering responses when the user is not engaged or present, leading to inefficient resource usage and power consumption.

Innovation Solution

A system that uses sensors and historical data to assess user engagement levels and environmental conditions to determine the optimal time for delivering AI-based voice responses, reducing unnecessary processing and power consumption by preempting responses when the user is not attentive.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the digital assistant continuously monitors and processes voice commands, then the responsiveness and availability of the assistant is improved, but the power consumption and resource usage increase

Engineering Contradiction:
Improveassistant availabilityVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary assessment of user engagement status before processing voice commands. By evaluating sensors data (audio, visual, contextual) in advance to determine whether the user is engaged, the system can preemptively avoid processing commands when the user is not attentive, thus reducing unnecessary power consumption while maintaining availability when needed

Inventive Principle:
Principle #10Preliminary action

2Speed

If the digital assistant delivers responses immediately upon receiving voice commands, then the responsiveness is improved, but the accuracy of response delivery deteriorates when the user is not engaged

Engineering Contradiction:
Improveresponse speedVSAvoidresponse accuracy
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The system incorporates feedback from multiple sensors (audio activity, visual engagement, contextual data) to continuously assess user engagement status. This feedback mechanism allows the assistant to determine in real-time whether the user is engaged before delivering responses, ensuring accuracy while maintaining speed through efficient decision-making based on sensor inputs

Inventive Principle:
Principle #23Feedback

3Productivity

If the digital assistant processes all voice commands without filtering, then the completeness of command processing is improved, but the efficiency and productivity deteriorate due to unnecessary processing

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidcommand completeness
Core Design Contradiction:
ProductivityVSLoss of information

Solution Approach 1:

The system performs preliminary filtering of voice commands based on user engagement assessment before processing. By evaluating whether the user is engaged through sensor data analysis in advance, the system can preemptively filter out commands when the user is not attentive, improving processing efficiency while maintaining completeness by still processing commands when engagement is detected

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11269591B2Artificial intelligence based response to a user based on engagement level
Publication Date: 2022.03.08 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US11269591B2 patent drawing
  • US11269591B2 patent drawing
  • US11269591B2 patent drawing

AI summary

Aspects of the present invention disclose a method for delivering an artificial intelligence-based response to a voice command to a user. The method includes one or more processors identifying an audio command received by a computing device. The method further includes determining a first engagement level of a user, wherein an engagement level corresponds to an attentiveness level of the user in relation to the computing device based at least in part on indications of activities of the user. The method further includes identifying a first set of conditions within an immediate operating environment of the computing device, wherein the first set of conditions indicate whether to deliver a voice response to the identified audio command. The method further includes determining whether to deliver the voice response to the identified audio command to the user based at least in part on the first engagement level and first set of conditions.