Live Stream Event Detection via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Live streaming systems often face inaccuracies in scheduling information due to events extending beyond their expected end times, leading to outdated channel guides and human intervention being necessary for data tagging, which is inefficient and unable to handle multiple streams simultaneously.
Innovation Solution
A system utilizing machine learning models to classify and tag live streaming content in real-time, identifying start and end locations of events within audio or video streams, allowing for automated data tagging and efficient communication of relevant content portions to user devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human intervention is used for data tagging of live streams, then tagging accuracy can be maintained, but productivity is reduced and the system cannot handle multiple streams simultaneously
Solution Approach 1:
The patent replaces the mechanical human tagging system with an automated machine learning-based detection system. The system uses audio and video analysis algorithms to automatically identify and tag events in live streams, eliminating the need for manual human intervention while maintaining tagging accuracy and enabling simultaneous processing of multiple streams.
Solution Approach 2:
The system enables self-service tagging by allowing the live stream content itself to provide the tagging information. The machine learning model analyzes the audio and video content directly to identify events, transitions, and key moments, allowing the content to tag itself without external human intervention.
2Loss of information
If channel guide information is updated in real-time, then information accuracy is improved, but device complexity increases
Solution Approach 1:
The patent introduces a machine learning-based event detection system as an intermediary between the live stream content and the channel guide information. This intermediary automatically analyzes the stream content and generates accurate timing information, which is then used to update the channel guide, maintaining information accuracy while managing system complexity through automation.
Solution Approach 2:
The system performs preliminary analysis of the live stream content using machine learning models to predict and identify upcoming events before they occur. This allows the channel guide to be proactively updated with accurate information about upcoming transitions and events, maintaining information accuracy without requiring complex real-time manual monitoring.
Data Source
AI summary
Embodiments of the present disclosure are directed to, among other things, a system that may receive a stream of data content that corresponds with a content classification. A machine learning (ML) model may be selected based at least in part on the metadata and/or content classification. The data content may be inputted to the ML model and an output from the ML model may indicate that data content is associated with the content classification, a start location, and an end location of the stream of data content. The stream of data content may be truncated or cut by removing portions of the stream of data content that fall before the start location and after the end location. The remaining portion of the stream may be stored as a file and provided to the user device.