Voice Assistant Authentication via Body Vibration Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current voice assistant systems lack robust security features, making them vulnerable to attacks such as replay, impersonation, and unauthorized access due to the open nature of voice commands, which can lead to information theft and financial harm.
Innovation Solution
A system that uses a wearable device with an accelerometer to record body vibrations associated with speech, combined with a microphone to record speech signals, and a multi-stage matching algorithm to authenticate voice commands by correlating the vibrations with the speech signals, providing continuous and secure authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice authentication mechanisms are implemented, then security is improved, but they fail to provide comprehensive security features and contain flaws
Solution Approach 1:
The authentication system is divided into multiple independent stages: initial voiceprint enrollment, continuous voice activity detection, and periodic voiceprint matching. Each stage operates independently to provide layered security, where the voiceprint module extracts features, the VAD module detects speech presence, and the matching module verifies authenticity, collectively addressing the limitation of single-mechanism authentication
Solution Approach 2:
The patent combines multiple authentication mechanisms into a composite security system that integrates voiceprint biometrics with voice activity detection. This composite approach merges the strengths of both methods—voiceprint provides unique identification while VAD ensures continuous verification—creating a more comprehensive and flaw-resistant security feature set
2Reliability
If voice assistant systems are made more secure, then security is improved, but device complexity increases
Solution Approach 1:
The system performs voiceprint enrollment and feature extraction in advance during an initial setup phase. The extracted voiceprint features are stored for future comparisons, eliminating the need for real-time complex processing during authentication. This preliminary preparation significantly reduces the computational burden and complexity during actual voice command execution
Solution Approach 2:
The patent introduces a voice activity detection module as an intermediary between the microphone input and the voiceprint matching process. This intermediary layer filters and pre-processes the audio stream, identifying speech segments that require authentication while discarding non-speech portions, thereby simplifying the overall system architecture and reducing the complexity of the authentication logic
3Reliability
If multiple authentication stages are implemented, then comprehensive security is improved, but processing time increases
Solution Approach 1:
The authentication system operates in periodic cycles: the voice activity detection module continuously monitors the audio stream at regular intervals, triggering voiceprint matching only when speech is detected. This periodic operation pattern allows the system to remain in a low-power, low-processing state during non-speech periods while providing comprehensive security verification when needed, thus balancing security thoroughness with processing time efficiency
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively prevents unauthorized access and voice command injection by ensuring that only genuine voice commands from the user are accepted, enhancing the security of voice assistant systems and protecting against various attack methods.
Implementation Method 1
an accelerometer device corresponding to a user and communicatively coupled to the one or more processors, wherein the accelerometer is configured to record vibrations corresponding to speech from the user
Implementation Method 2
a microphone communicatively coupled to the one or more processors, wherein the microphone is configured to record speech signals
Data Source
AI summary
A system and method for authenticating voice commands for a voice assistant includes an accelerometer device for recording vibrations corresponding to speech from a user. The recorded vibrations are compared to speech signals recorded by a microphone to determine if the speech signals originated from the user.