Voice Biometric Authentication Using Bone-Conducted Signal Verification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Voice biometric systems face challenges in differentiating between a user's speech and other audio sources, leading to potential false updates of stored voice prints and security risks in unsupervised enrichment processes.

Innovation Solution

The method involves obtaining both bone-conducted and air-conducted audio signals to determine when a user is speaking, enabling updates to the stored voice model only when the bone-conducted signal indicates a voice signal, thereby ensuring accurate differentiation and secure updates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If unsupervised enrichment is used to update voice prints using any detected speech, then the frequency of voice print updates increases and system performance improves, but the risk of erroneous updates using non-user speech increases leading to false rejections and security risks

Engineering Contradiction:

Improvefrequency of voice print updatesVSAvoidaccuracy of voice print updates

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The audio signal is segmented into bone-conducted and air-conducted components, allowing the system to separately analyze and identify user speech versus environmental speech. This segmentation enables selective enrichment using only authenticated user speech segments.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Bone-conducted audio signals serve as an intermediary verification mechanism. The system uses bone-conducted signals (which primarily capture user speech) to verify that air-conducted signals should be used for enrichment, acting as a mediator to ensure update accuracy.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Device complexity

If only air-conducted audio signals are used for voice print updates, then the system is simpler to implement, but the system cannot reliably differentiate between user speech and other audio sources

Engineering Contradiction:

Improvesimplicity of enrichment processVSAvoidaccuracy of speech source differentiation

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The system merges bone-conducted and air-conducted audio signal processing into a unified enrichment workflow. The bone-conducted signal processing provides verification capability while the air-conducted signal provides the actual enrichment data, combining strengths of both approaches.

Inventive Principle:

Principle #5Merging (Combining)

3Reliability

If supervised enrichment is used to update voice prints with user confirmation, then the accuracy of updates is high, but the user must actively participate which reduces convenience and increases time requirements

Engineering Contradiction:

Improveaccuracy of voice print updatesVSAvoiduser convenience in enrichment process

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The system performs self-service enrichment by automatically detecting and verifying user speech through bone-conducted signals, then autonomously using air-conducted signals for voice print updates without requiring user initiation or confirmation, making the process both accurate and convenient.

Inventive Principle:

Principle #25Self-service

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

This approach enhances the accuracy and security of voice biometric systems by ensuring that only the user's speech is used for updating voice prints, reducing false rejections and maintaining system performance.

Implementation Method 1

obtaining a first audio signal that comprises a representation of a bone-conducted signal, wherein the bone-conducted signal is conducted via at least part of the user's skeleton

Methodology Applied

Scientific EffectBone conduction:

Implementation Method 2

obtaining a second audio signal that comprises a representation of an air-conducted signal

Methodology Applied

Scientific EffectAir conduction:

Data Source

PatentUS10878825B2Biometric processes

Publication Date: 2020.12.29 CIRRUS LOGIC INC

AI summary

The present disclosure provides methods, systems, devices and computer program products for authenticating a user based on a comparison of audio signals to a stored voice model for an authorised user. In one aspect, a method comprises: obtaining a first audio signal that comprises a representation of a bone-conducted signal, wherein the bone-conducted signal is conducted via at least part of the user's skeleton; obtaining a second audio signal that comprises a representation of an air-conducted signal; and, responsive to a determination that the first audio signal comprises a voice signal, enabling updates to the stored voice model for the authorised user based on the second audio signal.