Voice Authentication Using Biometric Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice password authentication methods in electronic devices are insecure as voice passwords can be easily exposed, and users face inconvenience in remembering them, leading to a need for enhanced security and authentication speed.
Innovation Solution
An electronic device with a processor, memory, and input device that receives a voice command, acquires user identification information and voice print information, searches reference voice print information, and performs authentication based on the acquired voice print information, requesting additional speech if necessary to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If voice password authentication is used, then authentication speed is improved, but security deteriorates because voice passwords can be easily exposed
Solution Approach 1:
The authentication process is segmented into two independent components: voice print information (biometric characteristic) and identification information (semantic content). The voice print information is extracted from the voice signal and compared against stored templates, while the identification information is verified against pre-registered passwords. This segmentation allows the system to maintain high authentication speed through voice recognition while enhancing security by requiring both biometric match and correct identification, preventing simple voice replay attacks.
2Productivity
If voice password authentication is used, then authentication speed is improved, but user convenience deteriorates because users must remember voice passwords
Solution Approach 1:
The system performs self-service by automatically extracting voice print information from the user's voice command without requiring the user to consciously remember or recite a specific password. The identification information is obtained naturally from the voice command itself (e.g., the user's name or a natural phrase), eliminating the need for users to memorize complex voice passwords while maintaining fast authentication through automated voice print comparison.
3Measurement precision
If additional speech is requested to enhance accuracy, then authentication accuracy is improved, but authentication time increases
Solution Approach 1:
The system implements feedback by evaluating the quality and confidence level of the extracted voice print information and identification information. When the initial voice command provides sufficient clear voice print data and identification information with high confidence, the system completes authentication immediately. When the voice quality is poor or identification information is ambiguous, the system provides feedback by requesting additional speech, thereby improving accuracy only when necessary and minimizing authentication time for clear, confident cases.
Data Source
AI summary
Various embodiments of the present invention relate to a method and an electronic device for authenticating a user by using a voice command. Here, the electronic device may comprise a memory, an input apparatus, and a processor, wherein the processor is configured to: receive a voice command from the input apparatus; acquire user identification information and voice print information from the voice command; search reference voice print information of each of multiple users stored in the memory, for reference voice print information corresponding to the acquired user identification information; and perform authentication on the basis of the acquired voice print information and the reference voice print information. Other embodiments are also possible.