Voice Command Safety Analysis for Ambiguity Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice-enabled systems face recognition ambiguities due to the inability to directly invoke user-defined voice commands, leading to potential misinterpretation of commands like 'Brake' and 'Breakfast', resulting in unintended actions, such as a vehicle stopping abruptly.
Innovation Solution
A safety analysis is performed to assess the likelihood of confusion between new user-defined voice commands and existing commands, allowing direct invocation without a keyword if confusion is low, and requiring a keyword or restricting access if confusion is high, with options to rename or delete conflicting commands to mitigate ambiguity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If user-defined voice commands are allowed to be directly invoked without a keyword, then the ease of operation is improved, but the reliability deteriorates due to potential recognition ambiguities with system commands
Solution Approach 1:
The system performs a safety analysis before allowing direct invocation of user-defined commands. This preliminary check compares the user-defined command against system commands to assess confusion likelihood, and only permits direct invocation if the command is sufficiently distinct, thereby preventing recognition ambiguities before they can occur
Solution Approach 2:
The safety analysis engine acts as an intermediary between the speech recognition engine and the command execution system. It evaluates whether a user-defined command is likely to be confused with system commands and imposes appropriate invocation restrictions (direct or with keyword) based on this evaluation, mediating between ease of use and recognition accuracy
2Reliability
If a keyword is required before user-defined commands, then the reliability of command recognition is improved, but the ease of operation deteriorates due to additional utterance requirements
Solution Approach 1:
The system dynamically adjusts the invocation requirements for user-defined commands based on their similarity to system commands. Commands with low confusion likelihood permit direct invocation without keywords, while commands with high confusion likelihood require keywords. This dynamic adaptation optimizes both reliability and ease of operation for each specific command
3Reliability
If the system performs safety analysis on all user-defined commands, then the reliability of command recognition is improved, but the device complexity increases due to additional analysis requirements
Solution Approach 1:
The safety analysis functionality is extracted as a separate, dedicated engine that operates independently from the core speech recognition and command execution systems. This modular extraction allows the safety analysis to be performed without significantly complicating the main system architecture, as it functions as a distinct preprocessing layer
Data Source
AI summary
The present invention discloses a solution for assuring user-defined voice commands are unambiguous. The solution can include a step of identifying a user attempt to enter a user-defined voice command into a voice-enabled system. A safety analysis can be performed on the user-defined voice command to determine a likelihood that the user-defined voice command will be confused with preexisting voice commands recognized by the voice-enabled system. When a high likelihood of confusion is determined by the safety analysis, a notification can be presented that the user-defined voice command is subject to confusion. A user can then define a different voice command or can choose to continue to use the potentially confusing command, possibly subject to a system imposed confusion mitigating condition or action.