Speech Analyzing Unit for Voice Phishing Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Voice phishing attacks, which utilize VoIP features like caller ID spoofing and automated systems, are difficult to detect and prevent, especially for vulnerable individuals such as the elderly, as they can be personable and exploit trust, with existing solutions like STIR/SHAKEN and blacklisting approaches being insufficient in providing comprehensive protection.

Innovation Solution

A speech-analyzing-unit implemented as software or hardware within telecommunication devices, which automatically inspects call contexts to detect voice phishing patterns, translates patterns into multiple languages, and uses a service provider for further investigation, enabling efficient and timely protection across different languages and scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated text-to-speech systems and caller ID spoofing are used in voice phishing attacks, then the ability to conduct phishing attacks at scale is improved, but the detectability by law enforcement and victims deteriorates

Engineering Contradiction:
Improvephishing attack scaleVSAvoiddetectability
Core Design Contradiction:
ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent replaces manual analysis of voice phishing calls with an automated speech analyzing unit that uses speech recognition and pattern matching algorithms to detect phishing attempts. This automated system substitutes human effort with technological means, enabling scalable detection that matches the scaled-up phishing attacks.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements feedback by continuously analyzing speech patterns from incoming calls, comparing them against known phishing patterns, and providing real-time warnings to users. The system learns from new phishing attempts and updates its detection capabilities, creating a feedback loop that improves detection over time.

Inventive Principle:
Principle #23Feedback

2Adaptability or versatility

If speech pattern analysis is performed in multiple languages, then the coverage of phishing detection is improved, but the complexity of the detection system increases

Engineering Contradiction:
Improvelanguage coverageVSAvoiddetection system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The speech analyzing unit is designed with multi-functionality to handle multiple languages simultaneously. It incorporates speech recognition capabilities that can process and analyze speech patterns across different languages using a unified framework, allowing the same system to serve multiple linguistic contexts without requiring separate specialized systems for each language.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Speed

If the speech-analyzing-unit is implemented in telecommunication devices, then the speed of detection is improved, but the resource consumption of the device increases

Engineering Contradiction:
Improvedetection speedVSAvoiddevice energy consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The system applies partial action by not continuously analyzing all aspects of every call. Instead, it focuses on detecting specific speech patterns and keywords associated with phishing attempts, performing analysis only when suspicious patterns are detected. This selective approach reduces overall computational load and energy consumption while maintaining effective detection speed.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4156665B1Techniques to enable protection against voice phishing of a telephone call
Publication Date: 2024.10.30 DEUTSCHE TELEKOM AG
  • EP4156665B1 patent drawingFigure 1
  • EP4156665B1 patent drawingFigure 2~3
  • EP4156665B1 patent drawingFigure 4~5

AI summary

Techniques to enable protection against voice phishing of a telephone call performed by a user with a telecommunication device comprising the following steps: • Providing a speech-analyzing-unit that is assigned to the telecommunication device, wherein the speech-analyzing-unit comprises o an input data link to receive parts of the telephone call as audio input and/or to receive update information data; o a database configured to store speech patterns of possible voice phishing; o an algorithm configured to convert the audio input to text elements and to compare the text elements to the patterns of possible voice phishing with regard to an occurrence of voice phishing, and triggering countermeasures if voice phishing is detected; o at least an output data link to transmit the countermeasures to the telecommunication device and/or to a further user device.