Voice Transition Pattern Verification for Unproctored Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unproctored testing is vulnerable to cheating and lacks accuracy due to its susceptibility to impersonation, which undermines the fairness and reliability of assessment results, especially in cost-sensitive environments like employment and education.
Innovation Solution
A computer-implemented method that determines a user's transition pattern in voice data from responses to questions of varying difficulty levels, combining testing and verification processes to generate an integrated input, which is then stored for future verification, using features like pitch, formants, and glottal excitation, and optionally incorporating mouse and keystroke data to create a unique identity signature.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If unproctored testing is implemented to reduce costs, then cost savings are achieved, but security and reliability of test results deteriorate due to increased cheating and impersonation
Solution Approach 1:
The patent combines the testing process and verification process into a single integrated system. Voice data collected during the test is analyzed to create a transition pattern that serves as both the test record and the verification identifier, eliminating the need for separate proctored verification sessions while maintaining security and reliability
Solution Approach 2:
The patent introduces voice transition patterns as an intermediary mechanism that links the test-taker's identity to their test responses. This voice-based biometric marker acts as a mediator that verifies authenticity without requiring physical presence or additional security personnel
2Reliability
If separate verification processes are implemented to detect impersonation, then security improves, but user burden increases causing higher abandonment rates
Solution Approach 1:
The verification process is merged with the testing process by using the same voice data collected during test responses. The transition pattern analysis is performed on existing test data without requiring additional user actions, interviews, or separate verification sessions, thereby maintaining security while minimizing user burden
Solution Approach 2:
The system performs verification automatically using the voice data that the user naturally produces during testing. The transition pattern extraction and comparison are automated processes that require no additional effort from the user beyond providing their natural voice responses to test questions
Data Source
AI summary
Verifying a user, such as, but not limited to, a user who answered questions for an unproctered test for employment. A representation of a transition pattern is stored (200). Time series voice data is received from a user responding to a sequence of questions (72) having multiple levels of difficulty. The transition pattern in the voice data (200) is based on a plurality of comparisons (503) of subsets (500) of the voice data corresponding to responses to questions having one or more different levels of difficulty. A further transition pattern (200) is determined from voice data in the same way but based on a shorter sequence of questions (78). The user is verified based on a comparison (78) of this further transition pattern to a previously stored transition pattern.