Voice Input Authentication Device Using Signal Characteristic Learning Model

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing voice input authentication methods lack effectiveness in distinguishing between human voice and apparatus-generated voice, leading to potential security breaches.

Innovation Solution

A device and method utilizing a learning model to authenticate voice inputs by analyzing signal characteristic data, differentiating between voices uttered by humans and those produced by apparatuses, and employing additional authentication processes through voice patterns and query answering.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional voice input authentication is used, then the system is simple and easy to operate, but it cannot effectively distinguish between human voice and apparatus-generated voice leading to security breaches

Engineering Contradiction:

Improveauthentication reliabilityVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system performs preliminary extraction of signal characteristic data from voice inputs before authentication. This includes extracting features such as spectral centroid, spectral roll-off, and zero-crossing rate that are characteristic of human speech, creating a foundation for reliable distinction between human and apparatus-generated voices

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces traditional mechanical or rule-based voice authentication with a learning model-based system. The learning model automatically learns patterns from signal characteristic data to distinguish human voice from apparatus-generated voice, substituting manual authentication rules with adaptive machine learning

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Object-affected harmful factors

If voice input authentication without signal characteristic analysis is used, then the processing is fast and simple, but security against external attacks is insufficient

Engineering Contradiction:

Improvesecurity against attacksVSAvoidauthentication processing time

Core Design Contradiction:

Object-affected harmful factorsVSLoss of time

Solution Approach 1:

Signal characteristic data is extracted in advance from voice inputs before the authentication decision is made. This preliminary extraction of acoustic features enables the system to have security analysis ready when authentication is needed, reducing the time penalty during actual authentication

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The learning model acts as an intermediary between raw voice signals and authentication decisions. It processes signal characteristic data and provides security analysis, mediating between the need for fast processing and the need for thorough security checking

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11551699B2Voice input authentication device and method

Publication Date: 2023.01.10 SAMSUNG ELECTRONICS CO LTD

AI summary

Provided are a method of authenticating a voice input provided from a user and a method of detecting a voice input having a strong attack tendency. The voice input authentication method includes: receiving the voice input; obtaining, from the voice input, signal characteristic data representing signal characteristics of the voice input; and authenticating the voice input by applying the obtained signal characteristic data to a first learning model configured to determine an attribute of the voice input, wherein the first learning model is trained to determine the attribute of the voice input based on a voice uttered by a person and a voice output by an apparatus.