Inaudible Watermark for Virtual Partner Audio Source Identification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In public safety scenarios, it is difficult to differentiate between human voices, dispatch operators, and virtual partners in recorded audio, especially as audio quality improves, making it challenging to identify the source of audio evidence.

Innovation Solution

A method and apparatus are introduced to embed a unique watermark within virtual partner transmissions, using inaudible frequencies over 20 kHz, allowing for forensic identification of the virtual partner, officer, or their combination in recorded audio.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If audio quality is improved, then the clarity and fidelity of recorded audio is enhanced, but the ability to distinguish between human voices and virtual partner transmissions deteriorates

Engineering Contradiction:
Improveaudio qualityVSAvoidsource identification
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent introduces an inaudible watermark signal as an intermediary carrier of identification information. This watermark is embedded within the virtual partner's audio transmission and serves as a mediator that enables source identification without interfering with the audible audio quality. The watermark acts as a hidden layer of information that distinguishes virtual partner transmissions from human voices while maintaining the clarity and fidelity of the primary audio content.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If a watermark is embedded in the transmission, then the identification of virtual partner sources is enabled, but the complexity of the transmission system increases

Engineering Contradiction:
Improvesource identificationVSAvoidtransmission system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent employs parameter changes by modifying the audio signal in the frequency domain, specifically utilizing ultrasonic frequencies above 20 kHz that are inaudible to humans. By changing the frequency parameters of the transmission to include these high-frequency components, the system embeds identification information without affecting the audible range. This approach enables reliable source identification while minimizing the perceived complexity, as the changes occur in a frequency range imperceptible to human ears.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10587408B2Digital assistant water mark
Publication Date: 2020.03.10 MOTOROLA SOLUTIONS INC
  • US10587408B2 patent drawing
  • US10587408B2 patent drawing
  • US10587408B2 patent drawing

AI summary

A method and apparatus for providing a watermark within an audible virtual partner transmission is provided herein. During operation, each officer's digital partner will have a uniquely-identifiable watermark embedded within any audible transmission associated with the digital partner.