Voice Biometric System Configuration Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Voice biometric systems face challenges in accurately verifying users when the cohort of other speakers is not well-matched to the enrolled user's accent or audio conditions, leading to false determinations, especially when the enrolled user has a unique accent or different audio conditions during enrollment and verification.
Innovation Solution
The method involves comparing the received speech sample with a system configuration model, including a cohort and a Universal Background Model, using threshold values and score normalization parameters to adjust verification scores, and determining audio conditions matching the system configuration to ensure accurate user verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a cohort of other speakers is used for comparison during verification, then the ability to distinguish enrolled users from impostors is improved, but false determinations increase when the cohort is not well-matched to the enrolled user's accent or audio conditions
Solution Approach 1:
The system dynamically changes the parameters used for verification based on the configuration matching score. When the score indicates poor matching between the enrolled user and the cohort/UBM, the system adjusts the verification threshold or applies different normalization methods to account for the mismatch, thereby maintaining accurate similarity determination despite cohort limitations
Solution Approach 2:
The system performs a preliminary comparison of the enrolled user's speech characteristics with the cohort and UBM during the enrolment phase to generate a configuration matching score. This preliminary action identifies potential mismatches before verification begins, allowing the system to pre-adjust verification parameters to compensate for expected inaccuracies
2Reliability
If score normalization parameters are adjusted to account for configuration mismatches, then false acceptance and rejection rates are reduced, but the complexity of the verification process increases
Solution Approach 1:
The system automatically determines the configuration matching score and selects appropriate normalization methods without requiring manual intervention. The verification process self-adjusts based on the enrolled user's characteristics, reducing operational complexity while maintaining high reliability through automated parameter selection
3Reliability
If the system compares speech samples with both cohort models and UBM, then the verification process becomes more robust, but the computational time and resources increase
Solution Approach 1:
The system performs a partial comparison by focusing on the most discriminative features when computing the configuration matching score. Rather than exhaustively comparing all speech parameters with both cohort and UBM, the system identifies and processes only the critical configuration aspects, reducing computational time while maintaining verification robustness
Data Source
AI summary
A biometric system is tested to see whether a proposed use matches a configuration of the system. An enrolment input is received from an enrolling user, and compared with a system configuration model to obtain a configuration matching score value. The enrollment is then controlled based on a result of comparing the received enrollment input with the system configuration model. In the case of a voice biometric system, when a test input is received from a speaker, it is determined whether audio conditions applying to the test input correspond to system configuration conditions. Verification is performed by comparing the test input with a model of the speech of an enrolled user to generate a verification score for use in deciding whether to accept or reject the speaker, depending on whether it is determined that audio conditions applying to the test input correspond to the system configuration conditions.