Call Audio Background Filtering for Synthetic Sound Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to effectively detect synthetic sounds in call audio, which can be used maliciously to simulate urgent or sympathetic situations, leading to fraudulent activities.
Innovation Solution
A system that filters out voice audio to isolate background noise, analyzes the background noise for anomalies, and detects synthetic sound patterns by comparing against stored audio footprints, allowing for real-time mitigation of potential fraud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system analyzes all captured audio including voices, then detection accuracy improves, but processing time and resource consumption increase
Solution Approach 1:
The patent segments the audio signal into voice components and background audio layers using voice activity detection. By separating these components, the system can focus intensive analysis only on the background layer where synthetic sounds occur, rather than processing the entire audio stream including speech portions.
Solution Approach 2:
The patent extracts and removes voice components from the captured audio to isolate the background audio layer. This extraction process eliminates irrelevant voice data from subsequent analysis, reducing computational load while preserving the background sounds that contain the synthetic sound patterns to be detected.
2Speed
If the system processes background audio in real-time, then fraud detection responsiveness improves, but processing resources are consumed
Solution Approach 1:
The patent performs preliminary voice activity detection and audio segmentation before the main synthetic sound detection analysis. By pre-processing the audio to identify and isolate background portions, the system prepares the data in advance, enabling faster real-time detection without consuming excessive resources during the critical detection phase.
Solution Approach 2:
The patent applies partial action by focusing computational resources only on analyzing the background audio layer rather than the entire audio stream. This selective processing approach uses sufficient computation to detect synthetic sounds while avoiding the excessive resource consumption that would result from processing all audio components equally.
Data Source
AI summary
In some implementations, a system may capture audio from a call between a calling device and a called device. The system may filter the captured audio to generate a background audio layer. The system may generate an audio footprint that is a representation of sound in the background audio layer. The system may determine that the audio footprint includes a triggering sound footprint based on one or more audio characteristics of the audio footprint. The system may detect synthetic sound based on the audio footprint and after determining that the audio footprint includes the triggering sound footprint, wherein the synthetic sound is indicative of a sound recording. The system may transmit a notification to one or more devices associated with the call based on detecting the synthetic sound.