Decentralized Audio Fingerprint Matching for Scalable Audience Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio fingerprint recognition systems face high computational complexity and require centralized processing, making them inefficient for large-scale systems and lacking in scalability.
Innovation Solution
A decentralized audio fingerprint recognition system where portable devices perform matching using a customized algorithm with pre-known fingerprints, allowing for efficient distribution and scalability by tailoring the matching process for specific applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If centralized fingerprint processing is used, then processing accuracy is maintained, but system scalability and computational efficiency deteriorate
Solution Approach 1:
The patent segments the centralized fingerprint processing system into distributed processing units at individual devices. Each device performs fingerprint extraction and matching locally, dividing the overall system workload into independent segments that can operate simultaneously without requiring centralized coordination, thereby improving scalability and reducing system complexity.
Solution Approach 2:
The patent implements self-service by enabling each portable device to autonomously extract audio fingerprints and perform matching operations using locally stored reference fingerprints. The devices independently manage their own fingerprint processing without requiring external server intervention, which significantly improves processing efficiency and eliminates the bottleneck of centralized processing.
2Measurement precision
If audio fingerprint extraction is performed on all audio content, then identification accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies partial action by extracting and storing only the most salient acoustic features and temporal patterns necessary for accurate fingerprint identification, rather than processing every audio detail. This selective feature extraction maintains identification accuracy while significantly reducing the computational complexity required for fingerprint generation and matching.
Solution Approach 2:
The patent changes the parameter representation by transforming raw audio signals into compressed fingerprint representations that capture essential acoustic characteristics in a reduced dimensional space. This parameter transformation preserves identification accuracy while lowering computational requirements for subsequent matching operations.
3Speed
If fingerprint matching is performed locally on devices, then processing speed is improved, but device storage requirements increase
Solution Approach 1:
The patent extracts only the essential fingerprint information from the full audio content, separating the critical identification features from redundant audio data. By storing only these extracted fingerprint representations rather than complete audio files, the system achieves fast local matching speed while minimizing the storage capacity required at each device.
Solution Approach 2:
The patent creates compressed copies of audio fingerprints that contain sufficient information for accurate identification but occupy minimal storage space. These condensed fingerprint representations serve as efficient substitutes for storing complete audio content, enabling rapid local matching without requiring large storage capacities at individual devices.
Data Source
AI summary
Systems and methods are disclosed for customizing, distributing and processing audio signature data. Examples disclosed herein include waking a portable device from an inactive state in response to an activation signal. Disclosed examples also include generating a first signature based on audio from a microphone of the portable device during a period of time specified by the activation signal. Disclosed examples further include comparing the first signature with a second signature obtained from the activation signal to determine a match score, and communicating a match result based on the match score to a server via a network.