Audio Language Tagging for Faster Audience Measurement Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience measurement technologies face challenges in efficiently identifying and crediting media content as the list of media content and platforms grows, leading to increased computational resources and time required for language identification and matching.
Innovation Solution
Utilizing direct audio signals of media content to identify the language associated with the media content by tagging audio data with metadata, and creating meter events for language identifiers, which allows for efficient comparison with reference signatures within the same language database, reducing computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If media content identification is performed across all available content and platforms, then identification accuracy is improved, but computational resources and time required increase significantly
Solution Approach 1:
The patent segments the large reference database by organizing audio signatures according to their associated languages. This creates multiple smaller, language-specific database partitions that can be searched independently. The meter first identifies the language of captured audio content, then only searches within the corresponding language database segment, dramatically reducing the search space while maintaining complete coverage for accurate identification.
Solution Approach 2:
The patent performs preliminary language identification on captured audio content before proceeding to media content matching. This preliminary action filters the subsequent search process by pre-determining which language database to query. By establishing the language context in advance, the system avoids unnecessary comparisons across all languages, thereby reducing computational load while preserving identification accuracy.
2Measurement precision
If language identification is performed for all media content, then audience measurement accuracy is improved, but processing time increases
Solution Approach 1:
The reference database is segmented into language-specific subsets, allowing the system to perform targeted searches rather than exhaustive comparisons across all content. This segmentation enables parallel processing of language identification and content matching, reducing overall processing time while maintaining measurement accuracy through comprehensive language coverage.
Solution Approach 2:
Language identification is performed as a preliminary filtering step before content matching. This preliminary action reduces the effective search space by eliminating irrelevant language groups from consideration. The system achieves accurate audience measurement by ensuring the correct language database is queried, while processing time is reduced by avoiding unnecessary comparisons with unrelated language content.
Data Source
AI summary
In one example, a method is described. The method includes receiving, at a meter, audio data associated with media content from a media device; determining a language associated with the media content based on a media content voice analysis; assigning a language identifier associated with the language; and creating a meter event for the language identifier by tagging the audio data with metadata corresponding to the language identifier.