Audio Matching with Semantic Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audio processing techniques for audience measurement research fail to fully utilize semantic information and audio signature technology, relying on human classification and being overly complex, and do not effectively combine audio codes with semantic features to target advertisements based on listener preferences.
Innovation Solution
A processor-based method and system that reads audio codes and semantic audio signatures from media data, associating and processing them to provide supplemental information, including temporal, spectral, harmonic, and rhythmic features, to determine changing characteristics and generate targeted advertisements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If audio codes are used for audience measurement, then exposure detection is effective, but semantic information about media characteristics is lacking
Solution Approach 1:
The patent combines audio codes (for exposure detection) with semantic audio features (for media characteristics) into a unified measurement system. The audio signature system extracts temporal, spectral, harmonic, and rhythmic features from the audio signal, then merges these semantic features with the audio code data to create comprehensive media identification and classification capabilities.
Solution Approach 2:
The patent introduces semantic audio features as an intermediary between raw audio signals and high-level media classification. These features serve as a bridge that translates physical audio properties into meaningful characteristics like genre, instrumentation, and style, enabling both precise exposure detection and rich semantic analysis.
2Loss of information
If human classification is used for semantic analysis, then detailed characterization is achieved, but system complexity increases
Solution Approach 1:
The patent replaces manual human classification with an automated audio signature system that uses signal processing techniques. The system automatically extracts temporal, spectral, harmonic, and rhythmic features from audio signals and applies machine learning algorithms to classify media characteristics, eliminating the need for human analysts while maintaining comprehensive semantic analysis.
Solution Approach 2:
The audio signature system is self-organizing and automatically adapts to different media types without requiring manual configuration. The system autonomously extracts features, learns patterns from training data, and performs classification independently, reducing operational complexity while providing detailed semantic characterization.
3Reliability
If traditional audio matching is used, then code detection is reliable, but listener preference analysis is limited
Solution Approach 1:
The patent adds a new dimension of analysis by incorporating semantic audio features alongside traditional audio code matching. This creates a multi-dimensional measurement system that not only detects whether media was played (traditional dimension) but also characterizes the media properties and infers listener preferences (new semantic dimension), enabling targeted advertising based on actual listening behavior and preferences.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables effective audience measurement and targeted advertising by combining audio codes with semantic information, providing deeper insights into listener preferences and improving the accuracy of audience analysis.
Implementation Method 1
Known techniques exploit the psychoacoustic masking effect of the human auditory system whereby certain sounds are humanly imperceptible when received along with other sounds.
Data Source
AI summary
System, apparatus and method for determining semantic information from audio, where incoming audio is sampled and processed to extract audio features, including temporal, spectral, harmonic and rhythmic features. The extracted audio features are compared to stored audio templates that include ranges and/or values for certain features and are tagged for specific ranges and/or values. The semantic information may be associated with audio codes to determine changing characteristics of identified media during a time period.