Context-Aware Geo-Targeted Ad Delivery in Video Conferencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video conferencing services incur significant bandwidth costs, and existing advertising methods are ineffective in targeting users during live communication sessions without a keyboard interface, limiting the ability to display relevant geo-targeted advertisements.
Innovation Solution
The system captures audio from video or audio communication sessions, converts it into text, identifies relevant keywords, and combines this information with geolocation data to provide real-time targeted advertisements through a geo-targeting advertisement service, allowing for the display of context-aware ads during or after the communication session.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional advertising methods are used during video conferencing, then service providers can generate revenue to offset bandwidth costs, but the advertisements are not effectively targeted to users during live communication sessions
Solution Approach 1:
The patent replaces traditional mechanical input methods (keyboard typing) with voice-based audio input for keyword extraction. The system captures audio from the communication session, converts it to text, and automatically identifies keywords to trigger targeted advertisements, eliminating the need for manual user input while enabling precise advertising targeting during live video conferencing sessions
Solution Approach 2:
The patent introduces an intermediary processing system that acts as a mediator between the video conferencing audio stream and the advertising delivery system. This intermediary component performs audio-to-text conversion, keyword identification, and advertisement selection, bridging the gap between passive audio capture and active targeted advertising delivery without interrupting the live communication session
2Measurement precision
If keyboard interface is required for advertisement input, then precise advertising targeting can be achieved, but user interaction becomes complex and inconvenient during live communication
Solution Approach 1:
The patent substitutes the mechanical keyboard interface with an automated audio processing system. Voice commands and spoken keywords during the communication session are captured, transcribed to text, and automatically processed to identify advertising triggers, maintaining high precision in advertising targeting while completely eliminating the need for manual keyboard input during live communication
Solution Approach 2:
The system enables self-service advertising initiation by automatically extracting keywords from the communication session audio without requiring user intervention. The audio-to-text conversion and keyword identification processes occur autonomously, allowing the system to self-generate advertising triggers from natural conversation content, thereby simplifying user interaction while maintaining accurate advertising targeting
Data Source
AI summary
Techniques are provided for capturing audio containing speech at an endpoint device, converting the audio into corresponding text, identifying one or more keywords from the corresponding text and determining geolocation information associated with the endpoint device. Both the one or more keywords and the geolocation information are supplied as input to a geo-targeting advertisement service so as to cause the geo-targeting advertisement service to identify relevant content based on both the geolocation information and the one or more keywords. The geo-targeting advertisement service provides one or more advertisements to an endpoint device for display, wherein the advertisements are based on the one or more keywords and the geolocation information.