Virtual Assistant Gaze Detection Mode Switching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing virtual assistant technologies require a visual avatar for interaction, which consumes processing power and bandwidth, and do not allow seamless voice-based interaction when the user is not looking at the screen.

Innovation Solution

A method that switches between a displayed digital avatar and a voice-only virtual assistant based on user gaze, enabling full voice-based interaction without gestures or facial expressions when the user is not looking at the screen, thereby conserving processing power and bandwidth.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a visual digital avatar is displayed for virtual assistant interaction, then user engagement and interaction quality are improved, but processing power and bandwidth consumption increase

Engineering Contradiction:
Improveuser engagementVSAvoidprocessing power consumption
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

The system dynamically switches between two operational modes: a visual avatar mode when the user is looking at the screen, and a voice-only mode when the user is not looking. This dynamic adaptation allows the system to optimize resource consumption based on real-time user engagement state, resolving the contradiction between maintaining high user engagement and reducing processing power usage.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the operational parameters of the virtual assistant based on detected user gaze state. When the user is not looking at the screen, the system transitions from rendering visual avatar animations to audio-only processing, effectively changing the output modality parameter to reduce computational load while maintaining functional capability.

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If a visual digital avatar is displayed for virtual assistant interaction, then interaction quality is improved, but bandwidth usage increases

Engineering Contradiction:
Improveinteraction qualityVSAvoidbandwidth consumption
Core Design Contradiction:
Ease of operationVSLoss of energy

Solution Approach 1:

The system dynamically adjusts its output modality based on user gaze detection. When the user is not looking at the screen, the system switches from visual avatar rendering to audio-only responses, dynamically reducing bandwidth consumption for数据传输 while maintaining interaction functionality through voice output.

Inventive Principle:
Principle #15Dynamics

3Ease of operation

If the visual avatar is constantly displayed, then user engagement is maintained, but processing power is wasted when user is not looking at the screen

Engineering Contradiction:
Improveuser engagementVSAvoidprocessing efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The system uses its own sensor resources (camera/gaze detection) to monitor user engagement state and automatically adjusts its operational mode accordingly. This self-service mechanism allows the system to identify when visual display is unnecessary and switch to a lower-resource mode, improving processing efficiency without external intervention.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system implements a feedback loop where user gaze detection continuously monitors engagement state and feeds this information back to the virtual assistant module, which then adjusts its operational mode. This closed-loop feedback system ensures that processing resources are optimized based on real-time user attention state.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4080327B1Method involving virtual assistant
Publication Date: 2024.08.14 SALIENT WORLD AS
  • EP4080327B1 patent drawingFigure 1
  • EP4080327B1 patent drawingFigure 2
  • EP4080327B1 patent drawingFigure 3

AI summary

The present invention relates to a computer-implemented method, comprising: determining, using at least one sensor (24), whether or not a user (22) is looking at a computer program's window (26) on a screen (12) of a device (10); in response to determining that the user is looking at the computer program's window, displaying in the computer program's window a digital avatar (28) capable of interacting with the user with voice combined with at least one of gesture and facial expression; and in response to determining that the user is not looking at the computer program's window, enabling a voice-only virtual assistant (30) of the computer program (20) without displaying the digital avatar.