Multimodal Voice Authentication Using Bone Conduction and Gestures

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing voice authentication systems are susceptible to spoofing attacks and environmental noise, leading to unreliable user and device authentication.

Innovation Solution

Implementing a system that combines audio-related and bone conduction-related authentication modalities, using sensors to capture unique bone vibration and body gesture information, and employing AI-based models for enhanced authentication.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If voice authentication is used, then authentication can be performed remotely, but the system becomes susceptible to spoofing attacks and environmental noise

Engineering Contradiction:

Improveremote authentication capabilityVSAvoidauthentication security

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent combines multiple authentication modalities (audio-related, bone conduction-related, and body gesture-related) into a unified authentication system. The receiver component receives multiple types of authentication information simultaneously, and the authenticator component integrates these modalities to make authentication determinations, thereby maintaining remote accessibility while improving security through multi-factor verification

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces bone conduction as an intermediary channel between the audio signal and the authentication decision. By capturing bone vibration signals through the skull, the system creates an additional verification path that is resistant to spoofing attacks, as the bone conduction signal provides unique physiological information about the user's head and skull characteristics

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If multiple authentication modalities are combined, then authentication accuracy improves, but system complexity increases

Engineering Contradiction:

Improveauthentication accuracyVSAvoidsystem structure

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the authentication system into distinct functional components: a receiver component that captures different types of authentication information (audio, bone conduction, body gestures), an authenticator component that processes and compares the information, and a decision component that determines authentication status. This modular segmentation allows each component to specialize in specific modalities while maintaining overall system manageability

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The authenticator component is designed as a universal processor that handles multiple types of authentication information (audio-related, bone conduction-related, and body gesture-related) through a single integrated architecture. This multi-functional design enables the system to accommodate various authentication modalities without requiring separate dedicated processing systems for each modality

Inventive Principle:

Principle #6Universality (Multi-functionality)

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

Enhances authentication accuracy and security by reducing spoofing attacks and environmental noise interference, achieving performance comparable to face and fingerprint identification.

Implementation Method 1

at least one of bone conductive authentication information or body gesture-related authentication information

Methodology Applied

Scientific EffectBone conduction: Conduction (thermal)

Implementation Method 2

Enhanced sound-related and gesture-related authentication

Methodology Applied

Scientific EffectAcceleration detection: Accelerometer

Data Source

PatentUS20260030332A1Enhanced sound-related and gesture-related authentication

Publication Date: 2026.01.29 TDK USA CORP

AI summary

Enhanced sound-related and gesture-related authentication can be performed and managed. An authenticator can receive authentication information, comprising audio, bone conduction, and/or body gesture related authentication information, associated with an unauthenticated user from the unauthenticated user or an unauthenticated device associated with the unauthenticated user. Authenticator can determine whether to authenticate unauthenticated user and/or device based on analysis of the authentication information and stored authentication information, comprising stored audio, stored bone conduction, and/or stored body gesture related authentication information, associated with a verified user and/or verified device associated with verified user. If authenticator determines that authentication information satisfies defined match criteria with respect to stored authentication information, authenticator can determine unauthenticated user and/or device can be authenticated as the verified user and/or device. If authenticator determines the defined match criteria is not satisfied, authenticator can determine unauthenticated user and/or device cannot be authenticated.