Voice Biometric Authentication with One-Time Passcode
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current authentication methods, particularly in digital communication systems, face vulnerabilities such as spoofing attacks, including replay and synthesis attacks, which compromise the security of voice biometric systems, especially with the advancement of deepfake technologies.
Innovation Solution
The integration of voice biometric authentication with a one-time passcode, where a unique string is generated and verified through speech recognition and voice biometric analysis, combined with timing information comparison to distinguish between human and synthetic speech, providing an additional layer of security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If voice biometric authentication is used, then authentication convenience is improved, but security against spoofing attacks deteriorates
Solution Approach 1:
The patent combines voice biometric authentication with one-time passcode verification into a unified authentication system. The voice authentication module extracts biometric features from the user's voice, while simultaneously a one-time passcode is generated and verified. Both authentication methods must succeed for the transaction to be authorized, creating a merged security mechanism that maintains convenience while significantly improving resistance to spoofing attacks.
Solution Approach 2:
The authentication system uses a composite approach by integrating two different authentication mechanisms: voice biometrics (biological characteristic) and one-time passcodes (knowledge factor). This composite authentication method leverages the strengths of both approaches - the convenience of voice-based authentication and the security of time-limited passcodes - to create a robust security system that addresses the vulnerability of voice-only authentication to spoofing and deepfake attacks.
2Measurement precision
If voice biometric analysis is performed, then user verification capability is improved, but vulnerability to replay and synthesis attacks worsens
Solution Approach 1:
The system performs preliminary verification by requiring the user to read a dynamically generated one-time passcode during the voice authentication process. Before the voice biometric analysis can confirm user identity, the system first verifies that the spoken passcode matches the expected value. This preliminary action prevents replay attacks because recorded voice samples without the correct passcode will be rejected, and prevents synthesis attacks because generating the correct passcode in real-time is computationally infeasible for attackers.
Solution Approach 2:
The one-time passcode acts as an intermediary element between the user and the voice biometric verification system. Instead of directly verifying voice characteristics alone, the system uses the passcode as an intermediate verification step that must be successfully completed first. This intermediary mechanism adds a layer of security that prevents harmful factors like replay and synthesis attacks from compromising the underlying voice biometric data.
3Reliability
If one-time passcode with speech recognition is added, then security against spoofing is improved, but authentication complexity increases
Solution Approach 1:
The system employs speech recognition technology to automatically verify the one-time passcode without requiring manual input or additional user actions. The user simply speaks the passcode, and the speech recognition module automatically transcribes and validates it against the expected value. This self-service approach minimizes the perceived complexity for users while maintaining the security benefits of passcode verification, as the technical complexity of speech recognition is handled automatically by the system.
Data Source
AI summary
An authentication method, comprising:obtaining a first string;receiving an audio signal;performing speech recognition on the audio signal to extract text; andperforming authentication of a user, the authentication comprising:comparing the extracted text and the first string; andperforming voice authentication of the user based on the audio signal;generating a second string based on the first string and the extracted text;generating a data record using the second string; andstoring the data record in a data repository.