Voice Authentication Signal Harmonic Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing voice authentication systems require significant computing resources and struggle to accurately identify voices that change in volume or pitch, leading to increased processing time and data representation challenges, especially on mobile devices.

Innovation Solution

A method and system that generate voice identification by transforming voice signals into the frequency domain, setting predetermined amplitudes and frequencies for harmonics, and digitizing them, or by segmenting time domain signals into specific portions and digitizing their amplitudes and durations, to create a concise voice ID that can authenticate users across varying conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional voice authentication methods are used to extract relevant characteristics from voice signals, then voice identification accuracy is improved, but computational resource requirements increase significantly

Engineering Contradiction:
Improvevoice identification accuracyVSAvoidcomputational resource requirements
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the most essential voice characteristics (amplitude and frequency of harmonics) from the complete voice signal, discarding redundant information. This selective extraction maintains identification accuracy while significantly reducing computational requirements by focusing only on the critical features needed for authentication.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent transforms voice characteristics into simplified parameters by representing amplitudes with a first predetermined number of bits and harmonic counts with a second number of bits. This parameter transformation reduces the data complexity and computational burden while preserving the essential information needed for accurate voice identification.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If detailed voice characteristics are extracted to maintain identification accuracy under varying conditions, then authentication reliability is improved, but data representation size increases

Engineering Contradiction:
Improveauthentication reliabilityVSAvoiddata representation size
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent uses partial action by capturing only the essential voice characteristics (amplitude and frequency of harmonics) rather than the complete voice signal. This partial representation is sufficient to maintain authentication reliability across varying voice conditions while keeping the data representation compact and efficient.

Inventive Principle:
Principle #16Partial or excessive action

3Adaptability or versatility

If traditional voice processing methods are applied to handle voice variations, then adaptability to different environments is improved, but processing time increases

Engineering Contradiction:
Improveadaptability to voice variationsVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent performs preliminary normalization by setting the amplitude of the first harmonic to a predetermined value and adjusting other harmonics to maintain relative gain. This preliminary processing step standardizes the voice data before analysis, enabling faster processing while maintaining adaptability to voice variations in volume and pitch.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2337022B1System and method for low overhead voice authentication
Publication Date: 2011.12.28 BLACKBERRY LTD
  • EP2337022B1 patent drawingFigure 1
  • EP2337022B1 patent drawingFigure 2
  • EP2337022B1 patent drawingFigure 3

AI summary

A system and method are provided to authenticate a voice in a frequency domain. A voice in the time domain is transformed to a signal in the frequency domain. The first harmonic is set to a predetermined frequency and the other harmonic components are equalized. Similarly, the amplitude of the first harmonic is set to a predetermined amplitude, and the harmonic components are also equalized. The voice signal is then filtered. The amplitudes of each of the harmonic components are then digitized into bits to form at least part of a voice ID. In another system and method, a voice is authenticated in a time domain. The initial rise time, initial fall time, second rise time, second fall time and final oscillation time are digitized into bits to form at least part of a voice ID. The voice IDs are used to authenticate a user's voice.