Speaker Authentication Correction Models for Network Condition Mismatches
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing speaker authentication technologies face challenges in accurately verifying identities due to mismatches between enrollment and test process conditions in digital communication networks, particularly in mobile or IP networks, where varying hardware and network conditions lead to differences in voice characterizing parameter sets, resulting in incorrect authentication.
Innovation Solution
The implementation of correction parameters or correction models that normalize or adjust the test and enrollment voice characterizing parameter sets to standard conditions, compensating for mismatches caused by differences in hardware and network conditions, ensuring accurate authentication by aligning both sets to similar process conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If speaker authentication is implemented using voice characterizing parameter sets, then security verification capability is improved, but authentication accuracy deteriorates due to mismatches between enrollment and test process conditions
Solution Approach 1:
The patent applies parameter changes by transforming voice characterizing parameter sets through correction models that adjust acoustic features based on detected process conditions. The system changes the parameters of voice representations to compensate for environmental variations, hardware differences, and network conditions, thereby maintaining authentication accuracy across different enrollment and test scenarios.
Solution Approach 2:
The patent introduces correction models as intermediary components between the voice authentication system and the varying process conditions. These correction models act as mediators that detect mismatches in process conditions and apply appropriate corrections to voice characterizing parameter sets, resolving the conflict between security verification capability and authentication accuracy.
2Measurement precision
If correction parameters are applied to normalize voice characterizing parameter sets, then authentication accuracy is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-training correction models during the enrollment phase to learn the relationship between process conditions and voice parameter variations. This preliminary action allows the system to store correction parameters in advance, which are then applied during authentication without requiring complex real-time analysis, thus improving accuracy while managing system complexity.
Solution Approach 2:
The system applies self-service by enabling the correction models to automatically detect process condition mismatches and apply appropriate corrections without manual intervention. The authentication system self-adjusts by using the correction parameters to normalize voice characterizing parameter sets, reducing the need for complex external processing while maintaining high authentication accuracy.
Data Source
AI summary
Example embodiments provide a speaker authentication technology that compensates for mismatches between enrollment process conditions and test process conditions using correction parameters or correction models, which allow for correcting one of the test voice characterizing parameter set and the enrollment voice characterizing parameter set according to a mismatch between the test process conditions and the enrollment process conditions, thereby obtaining values for the test voice characterizing parameter set and the enrollment voice characterizing parameter set that are based on the same or at least similar process conditions. Alternatively, each of the enrollment and test voice characterizing parameter sets may be normalized to predetermined standard process conditions by using the correction parameters or correction models.