Modulated Signal Insertion for Multiparty Call Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems for detecting multiparty calls in controlled-environment facilities face challenges such as false positives due to limitations in voice recognition, inability to differentiate multiple speakers, and the need for extensive data sharing across jurisdictions, which complicates the identification of resident-to-resident communications.

Innovation Solution

The introduction of a modulated signal into the audio of a controlled-environment facility resident's call, which can be detected by a communication apparatus to determine if multiple residents are participating in a call, providing real-time identification and alerting authorities to the presence of multiple residents on the same call.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If voice recognition-based multiparty call detection is used, then communication monitoring capability is improved, but false positive rate increases due to inability to differentiate multiple speakers

Engineering Contradiction:

Improvemultiparty call detection capabilityVSAvoiddetection accuracy

Core Design Contradiction:

Difficulty of detecting and measuringVSReliability

Solution Approach 1:

The system segments the audio signal by creating separate audio streams for each phone line, allowing independent analysis of each call path. This enables the system to track which specific resident voices appear in which calls, preventing false positives from concurrent legitimate calls.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces call detail records as an intermediary data structure that captures voiceprint information, call timing, and phone line associations. These records serve as a mediator between the audio analysis and the multiparty detection logic, enabling accurate tracking of resident participation across multiple calls.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Speed

If real-time multiparty call detection is implemented, then monitoring responsiveness is improved, but system complexity increases due to concurrent call analysis requirements

Engineering Contradiction:

Improvedetection response timeVSAvoidsystem architecture complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by continuously capturing and storing call detail records with voiceprint information during call setup and progression. This pre-processing allows the multiparty detection logic to operate on already-structured data, reducing computational complexity during real-time analysis.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms where call detail records are continuously updated and fed back into the detection system. This allows the system to maintain an evolving understanding of active calls and resident participation, enabling real-time detection without requiring complete re-analysis of all calls.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If extensive data sharing across jurisdictions is required for voice modeling, then detection accuracy is improved, but data sharing complexity and privacy concerns increase

Engineering Contradiction:

Improvevoiceprint matching accuracyVSAvoiddata sharing infrastructure complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system creates a universal voiceprint identification framework that can operate within a single jurisdiction or across multiple jurisdictions. The call detail record structure and voiceprint matching algorithms are designed to be jurisdiction-agnostic, allowing the same system to function whether sharing data across facilities or operating independently.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Measurement precision

If post-call voice modeling and comparison is used, then detection thoroughness is improved, but real-time monitoring capability deteriorates

Engineering Contradiction:

Improvevoice comparison accuracyVSAvoiddetection delay

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system maintains continuous useful action by performing both real-time detection using call detail records and post-call voice modeling. The call detail records enable immediate detection during calls, while voice modeling continues in the background for enhanced accuracy, with results fed back to improve future real-time detection.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS10142462B2Controlled-environment facility resident communication detection

Publication Date: 2018.11.27 SECURUS TECH LLC

AI summary

Systems and methods insert a modulated signal into audio of a controlled-environment facility resident's side of a call, such as by a controlled-environment communication apparatus, or the like. The same or a different controlled-environment communication apparatus, or the like, may detect the same and/or another modulated signal inserted into the audio of the same and/or another controlled-environment facility resident's side of the call in the audio of an other party's side of the call. The same or a different controlled-environment communication apparatus, or the like, may determine if more than one modulated signal is present in the other call party's side of the call and alert an authority that more than one controlled-environment facility resident is taking part in the call.