Digital Video Recorder Commercial Skipping via Caption Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital video recorders (DVRs) require manual user intervention to skip commercials, leading to inaccuracies as users may fast-forward past desired content or resume playback in unwanted sections, due to the lack of automated segmentation and volume adjustment capabilities for distinguishing between program content and interstitials.
Innovation Solution
A method and apparatus that process a presentation stream by using closed captioning data to identify segment boundaries and adjust audio/video frame volumes, allowing for automatic skipping of commercials and adjustment of playback rates, based on location information and signatures within the stream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual fast forwarding is used to skip commercials, then the user can control playback, but the user may fast forward past desired content or resume playback during unwanted sections
Solution Approach 1:
The system automatically identifies and processes commercial segments without requiring user intervention. The DVR analyzes the recorded content, detects commercial boundaries using various methods (captioning data, audio analysis, video analysis), and automatically fast-forwards through identified commercial portions during playback, allowing the content to 'serve itself' by eliminating the need for manual user control during commercial skipping
Solution Approach 2:
The system uses multiple feedback mechanisms to improve commercial detection accuracy: analyzing closed captioning data for commercial indicators, monitoring audio level changes that distinguish commercials from program content, examining video frame differences, and using user feedback when manual verification is required. This multi-layered feedback approach continuously refines the accuracy of commercial identification and boundary detection
2Measurement precision
If automated commercial skipping is implemented, then accuracy of skipping improves, but device complexity increases
Solution Approach 1:
The commercial skipping system is divided into separate functional modules: a captioning data analysis module that processes closed captioning for commercial indicators, an audio analysis module that detects commercial boundaries through audio level monitoring, a video analysis module that examines frame differences, and a control module that coordinates fast-forward operations. This segmentation allows each module to specialize in specific detection tasks, improving overall accuracy while managing system complexity through modular design
Solution Approach 2:
The system employs multiple detection methods (captioning analysis, audio monitoring, video frame comparison) beyond what a single method could provide. By using excessive action - applying multiple overlapping detection techniques - the system achieves higher accuracy in commercial identification. The control logic selectively applies these methods based on confidence levels, using partial action from each method while combining them for comprehensive detection
3Measurement precision
If multiple detection methods are used to identify commercial boundaries, then identification accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary analysis of recorded content during the recording phase, pre-identifying potential commercial segments and storing this information for later use. Closed captioning data is analyzed and stored with timestamps, audio characteristics are pre-processed and saved, and video frames are pre-compared to identify potential commercial boundaries. This preliminary action reduces the processing burden during playback, as the system can quickly retrieve pre-analyzed data rather than performing full analysis in real-time
Solution Approach 2:
The system implements periodic sampling of audio and video data during commercial detection, analyzing frames at intervals rather than continuously. Audio levels are monitored at regular sampling rates, and video frame comparisons are performed periodically rather than on every frame. This periodic action significantly reduces processing time while maintaining adequate detection accuracy, as the system captures sufficient information at sampled intervals to identify commercial boundaries without the computational burden of continuous analysis
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Described herein are methods and apparatus for the identification of locations in a presentation stream (104) based on metadata associated with the presentation stream. Locations within a presentation stream are identified using location information (108) and/or signature data (106) associated with the presentation stream. The identified locations within a presentation stream may be utilized to identify boundaries (208, 210) of segments (202, 206) within the presentation stream, such as segments of a show. The boundaries of the segments may then be utilized to change the volume of segments or playback a segment at a non-real time presentation rate (1518).