Voice Assistant Feedback Loop for Misrecognition Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic apparatuses equipped with voice assistant platforms often fail to recognize user voice commands accurately, leading to inconsistent or non-responsive operations, making it difficult for users to correct malfunctions and understand which commands are supported.
Innovation Solution
An electronic apparatus with a processor, microphone, inputter, and memory that receives user voice commands, stores operation information, and provides guide information based on user responses to ensure accurate command execution, including identifying and correcting misrecognized voice commands by displaying relevant messages or items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the electronic apparatus uses voice recognition to perform operations, then the ease of operation is improved, but the reliability deteriorates because the voice command recognition is not perfect (cannot recognize 100% accurately)
Solution Approach 1:
The system monitors user responses to operations performed based on voice commands. When a user provides feedback (either through voice or inputter) indicating the operation was incorrect or unwanted, the system stores this feedback information and uses it to adjust future voice recognition attempts, thereby improving reliability while maintaining ease of operation.
Solution Approach 2:
The system performs preliminary monitoring of user responses before finalizing the operation execution. By detecting user feedback within a threshold time period after voice command processing, the system can preemptively correct potential misrecognitions before they cause harmful effects, thus improving reliability without compromising the convenience of voice-based operation.
2Productivity
If the electronic apparatus performs operations based on voice commands without verification, then the productivity is improved, but the harmful factors increase because incorrect operations may be executed
Solution Approach 1:
The system implements a feedback mechanism that detects whether the performed operation matches the user's intended command. When mismatch is detected (user provides feedback indicating incorrect operation), the system can undo or correct the operation, thereby preventing harmful effects while maintaining high productivity through rapid voice-command response.
Solution Approach 2:
The system prepares for potential errors by monitoring user responses within a threshold time window. This beforehand cushioning allows the system to detect and correct incorrect operations before they cause harm, enabling the system to maintain high productivity while protecting against harmful effects of voice recognition errors.
3Device complexity
If the electronic apparatus does not provide feedback about its operations, then the device complexity is reduced, but the loss of information increases because users cannot understand which commands are supported or why operations failed
Solution Approach 1:
The system provides feedback to users about the operations performed based on voice commands and about supported commands. This feedback can be delivered through the display or speaker, helping users understand which commands are supported and why certain operations may have failed, thereby reducing information loss without significantly increasing device complexity.
Solution Approach 2:
The system uses the display and speaker as intermediary channels to communicate operation status and supported command information to users. These existing components serve as mediators that convey necessary information about voice command recognition and execution, reducing information loss without requiring additional complex feedback mechanisms.
Data Source
AI summary
An electronic device to perform an operation corresponding to a first user voice when the first user voice is received through the microphone; store, information about an operation corresponding to the first user voice and user reaction information including the user command when a user command is received through the input unit within the first threshold time from when the first user voice is received, or from when the operation corresponding to the first user voice is performed; perform an operation corresponding to a second user voice when the second user voice is received through the microphone; and provide guide information corresponding to the user command on the basis of the user reaction information stored in the memory when the type of operation corresponding to the first user voice is the same as the type of operation corresponding to the second user voice.