Voice Authentication After Non-Voice Identity Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice authentication systems face high processing loads when continuously verifying user identity during operation, as they compare voice features with all users, making it inefficient to prevent illicit operations by impersonators.
Innovation Solution
An information processing apparatus that performs initial authentication using a method other than voice, such as an IC card, and subsequent voice authentication by comparing the registered voice features of the user with the voice used during operation, reducing the processing load and time required for verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice authentication is performed by comparing a feature of the user's voice with features of voices of all users, then authentication accuracy is improved, but processing load becomes too heavy
Solution Approach 1:
The authentication process is divided into two distinct phases: initial authentication using non-voice methods (IC card, fingerprint, etc.) and subsequent voice authentication. This segmentation allows the system to use different authentication methods at different times, reducing the processing load during voice authentication by only comparing against the pre-identified user's voice features rather than all users.
Solution Approach 2:
The system performs preliminary authentication using non-voice methods before voice authentication. This preliminary action identifies the user in advance, so that when voice authentication is needed, the system only needs to verify the voice against the already-identified user's registered voice features, significantly reducing the comparison scope and processing load.
2Reliability
If voice authentication is performed continuously during operation, then security against illicit operations is improved, but processing time increases
Solution Approach 1:
Instead of continuous voice authentication, the system uses periodic authentication at specific moments: initial authentication before operation begins and subsequent authentication when operations are performed. This periodic approach maintains security while minimizing the time spent on authentication processes.
Solution Approach 2:
The system uses the user's registered voice features (a copy of their voice characteristics) for authentication comparisons rather than requiring continuous analysis of all user voice databases. This allows rapid verification against stored voiceprints, reducing processing time while maintaining security.
3Productivity
If only initial authentication is performed, then processing speed is improved, but security against impersonation deteriorates
Solution Approach 1:
The system introduces a intermediate step: after initial non-voice authentication identifies the user, the user's registered voice features are retrieved as an intermediary reference. This intermediary allows subsequent voice authentication to be performed quickly against this specific user's voiceprint rather than searching through all users, maintaining both speed and security.
Data Source
AI summary
An information processing apparatus includes a processor configured to obtain a registered first voice of a first user identified in first authentication achieved by an authentication method different from an authentication method employing a voice, obtain a second voice uttered by a second user during operation, perform second authentication for confirming that the second user is the first user identified in the first authentication by comparing a feature of the first voice and a feature of the second voice, and perform, if the second authentication is successfully completed, a process in accordance with an operation associated with the second voice.