Audio Thumbnail Fingerprinting for TV Metadata Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The migration to high-definition television and interactive television technologies is hindered by televisions being relegated to 'dumb' monitors, as they often lack metadata necessary for an enhanced experience, which is typically not supplied by set-top boxes, necessitating the ability for display devices to recover metadata from content itself.
Innovation Solution
A system and method that uses audio information to identify content by comparing audio thumbnails with pre-stored thumbnails, enabling the retrieval and display of metadata associated with the content, allowing display devices to render additional information and provide an interactive experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If televisions are used as 'dumb' monitors without set-top box assistance, then device complexity is reduced, but the ability to access metadata is lost
Solution Approach 1:
The television performs audio fingerprinting and metadata retrieval independently without requiring set-top box assistance. The TV extracts audio features from received content, generates fingerprints, queries external databases, and retrieves metadata autonomously, enabling self-service operation.
Solution Approach 2:
An external database server acts as an intermediary between the television and metadata. The TV sends audio fingerprints to the database, which returns corresponding metadata, enabling the TV to access information it cannot obtain directly from the broadcast signal.
2Loss of information
If set-top boxes supply metadata, then metadata availability is improved, but device complexity increases
Solution Approach 1:
The television performs audio fingerprinting and metadata retrieval independently without requiring set-top box assistance. The TV extracts audio features from received content, generates fingerprints, queries external databases, and retrieves metadata autonomously, enabling self-service operation.
Solution Approach 2:
The metadata retrieval function is extracted from the set-top box and implemented directly in the television. This separates the metadata acquisition capability from the conditional set-top box requirement, allowing the TV to obtain metadata independently.
3Adaptability or versatility
If audio fingerprinting is implemented in display devices, then metadata retrieval capability is improved, but processing requirements increase
Solution Approach 1:
The system extracts only essential audio features (frequency spectrum, temporal envelope) rather than processing the entire audio signal. Audio fingerprints are generated from selected time-frequency representations, reducing computational load while maintaining identification accuracy.
Solution Approach 2:
The television extracts only the necessary audio features for fingerprinting from the full audio signal, rather than processing all audio data. This selective extraction reduces processing requirements while enabling metadata retrieval functionality.
Data Source
AI summary
Herein, a display device, system and method are described for identifying digital content through comparison of one or more audio thumbnails when the digital content is provided without metadata. According to one embodiment, a display device comprises a display, audio conversion logic and a transceiver. The audio conversion logic is adapted to produce one or more audio thumbnails, which are compact representations of a snippet of audio from that incoming digital content without the metadata. The transceiver is adapted to transmit the one or more audio thumbnails in order to (i) retrieve metadata associated with the digital content from an external source separate from a source of the digital content and (ii) display information provided by the metadata on the display.