Speech Authentication Threshold Adjustment via Environmental Noise Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing user authentication methods using speech recognition face performance degradation due to environmental changes, such as noise and distance variations between registration and authentication sessions.

Innovation Solution

A method and apparatus that compare and adjust authentication criteria by generating environment information from input audio signals, distinguishing utterance and non-utterance sections, and matching them with registration environment information to adjust thresholds dynamically.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If speech recognition is performed using fixed authentication criteria registered in a controlled environment, then authentication accuracy is maintained under ideal conditions, but authentication performance degrades when environmental conditions change (noise, distance, reverberation)

Engineering Contradiction:
Improvespeech recognition accuracyVSAvoidauthentication performance consistency
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies dynamics by making the authentication criterion (threshold) variable rather than fixed. The threshold is dynamically adjusted based on environmental conditions detected during authentication. The system calculates a threshold adjustment value based on the difference between current environmental features (noise, reverberation, distance) and registered environmental features, then applies this adjustment to the original threshold. This allows the authentication system to adapt to changing environmental conditions while maintaining consistent performance.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of the authentication threshold based on environmental parameters. By detecting environmental features (noise level, reverberation characteristics, distance estimates) and comparing them with registered environmental features, the system calculates an adjustment value that modifies the authentication threshold. This parameter change approach allows the system to compensate for environmental variations and maintain reliable authentication performance across different conditions.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If the authentication threshold is lowered to accommodate environmental variations, then authentication reliability improves in changing environments, but false acceptance rate increases

Engineering Contradiction:
Improveauthentication performance consistencyVSAvoidauthentication accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent implements feedback by using environmental feature detection and comparison to generate a threshold adjustment value. The system continuously monitors environmental conditions, compares them with registered environmental features, and feeds this information back to adjust the authentication threshold accordingly. This feedback mechanism ensures that the threshold is neither too strict nor too lenient, maintaining both reliability and accuracy by adapting to actual environmental conditions rather than using fixed or uniformly lowered thresholds.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3850509B1User authentication method and apparatus
Publication Date: 2025.01.01 SAMSUNG ELECTRONICS CO LTD
  • EP3850509B1 patent drawingFigure 1
  • EP3850509B1 patent drawingFigure 2
  • EP3850509B1 patent drawingFigure 3~4

AI summary

A method and apparatus for authenticating a user based on an utterance input includes obtaining an input audio signal based on the utterance input of the user; obtaining, from the input audio signal, at least one audio signal of an utterance section and at least one audio signal of a non-utterance section; generating environment information indicating an environment in which the utterance input is received, based on the at least one audio signal of the non-utterance section; obtaining a result of a comparison between the generated environment information and registration environment information indicating an environment in which a registration utterance input corresponding to a previously registered registration audio signal corresponding to the user is received; adjusting an authentication criterion for authenticating the user based on the result of the comparison; and authenticating the user based on the adjusted authentication criterion and the input audio signal.