Voice Dialog Device Using Sight Line Detection to Reduce False Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice dialog systems face challenges in accurately determining the start and end of a dialog, leading to false responses and reduced convenience, as they often rely solely on voice recognition or physical triggers, which can result in unintended interactions.
Innovation Solution
A voice dialog device that combines sight line detection and voice processing to determine user intent, using both the direction of the user's gaze and voice recognition to initiate or terminate a dialog, thereby reducing false responses and enhancing user convenience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice recognition alone is used to determine dialog start, then the system is simple to operate, but false recognition occurs when keywords are mentioned in unrelated conversations
Solution Approach 1:
The patent combines sight line detection and voice recognition into a unified dialog determination system. The dialog determination unit integrates results from both the sight line detection unit and voice processing unit to determine whether to start a dialog, thereby reducing false recognition while maintaining reasonable system complexity through functional integration.
Solution Approach 2:
The dialog determination unit serves as an intermediary that processes and integrates information from both sight line detection and voice recognition systems. It mediates between the two detection methods by requiring both conditions to be met before initiating a dialog, thus improving reliability without directly increasing overall system complexity.
2Reliability
If physical contact trigger is used to start dialog, then false recognition is reduced, but user convenience is reduced due to required physical action
Solution Approach 1:
The patent replaces the mechanical physical contact trigger with a non-contact method combining sight line detection and voice recognition. The system detects the user's gaze direction and voice simultaneously, eliminating the need for physical contact while maintaining high accuracy in determining dialog start intent.
3Measurement precision
If sight line detection combined with voice processing is used, then dialog start accuracy is improved, but device complexity increases
Solution Approach 1:
The patent merges sight line detection and voice processing functions into an integrated system where the dialog determination unit coordinates both detection methods. This merging approach improves measurement precision for user intent detection while managing device complexity through unified control logic and shared processing resources.
Data Source
AI summary
A voice dialog device includes a sight line detection unit configured to detect a sight line of a user, a voice acquiring unit configured to acquire voice pronounced by the user, and a processor. The processor is configured to perform a step of acquiring a result of recognizing the voice, a step of determining whether or not the user is driving, and a step of determining whether or not the voice dialog device has a dialog with the user. When the detected sight line of the user is in a certain direction, and a start keyword has been detected from the voice, the processor determines that the user has started a dialog. The processor switches the certain direction based on whether the user is driving.