Speech Credit Segment Detection Using Overlapping Audio Clips
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying credit segments in broadcast content, such as sponsor credits, is inefficient due to their relatively short duration, requiring significant time and effort.
Innovation Solution
A credit segment identifying device that extracts and identifies credit segments using partial speech signals and associates them with second speech or video signals to determine the presence of credit segments through trained identifiers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual inspection of broadcast content is used to identify sponsor credit segments, then identification accuracy can be maintained, but time consumption increases significantly
Solution Approach 1:
The broadcast content is segmented into multiple short clips based on temporal boundaries. The extracting unit divides the continuous broadcast signal into discrete segments, which are then processed independently by the identification unit. This segmentation allows parallel processing and reduces the overall time required for identification while maintaining accuracy through focused analysis of each segment.
Solution Approach 2:
The manual visual inspection process is replaced with an automated identification unit that uses machine learning models to detect sponsor credit segments. The identification unit processes broadcast content through trained neural networks, substituting human visual inspection with automated computational analysis that operates faster and consistently without fatigue.
2Reliability
If the entire broadcast is processed to ensure no credit segments are missed, then identification completeness is improved, but processing efficiency decreases
Solution Approach 1:
The extracting unit performs preliminary processing by identifying and extracting potential credit segment clips before the main identification process. This preliminary action prepares the data in advance, allowing the identification unit to focus only on relevant segments rather than processing the entire broadcast, thus improving both completeness and efficiency.
Solution Approach 2:
The system maintains continuous processing of broadcast content through overlapping time windows and sequential analysis of extracted clips. The identification unit continuously analyzes extracted segments in a streamlined pipeline, ensuring no credit segments are missed while maintaining high processing throughput through optimized continuous operation.
Data Source
AI summary
A credit segment identifying device includes an extracting unit. The extracting units extracts, from a first speech signal, a plurality of first partial speech signals. Each of the plurality of first partial speech signals is a part of the first speech signals and shifted from each other in time direction. An identifying unit identifies a credit segment in the first speech signal by determining whether each of the first partial speech signals includes a credit according to an association between each of second partial signals and the presence/absence of a credit. The each of second partial signals is extracted from a second speech signal.


