CONF vinciarelli:icmevincia:2007/IDIAP Semantic Segmentation of Radio Programs Using Social Network Analysis and Duration Distribution Modeling Vinciarelli, Alessandro Fernàndez, F. Favre, Sarah EXTERNAL https://publications.idiap.ch/attachments/papers/2007/vinciarelli-icmevincia-2007.pdf PUBLIC https://publications.idiap.ch/index.php/publications/showcite/vinciarelli:rr06-75 Related documents IEEE International Conference on Multimedia and Expo (ICME) 2007 779-782 IDIAP-RR 06-75 This work presents and compare two approaches for the semantic segmentation of broadcast news: the first is based on Social Network Analysis, the second is based on Poisson Stochastic Processes. The experiments are performed over 27 hours of material: preliminary results are obtained by addressing the problem of splitting different episodes of the same program into two parts corresponding to a news bulletin and a talk-show respectively. The results show that the transition point between the two parts can be detected with an average error of around three minutes, i.e. roughly 5 percent of each episode duration. REPORT vinciarelli:rr06-75/IDIAP Semantic Segmentation of Radio Programs Using Social Network Analysis and Duration Distribution Modeling Vinciarelli, Alessandro Fernàndez, F. Favre, Sarah EXTERNAL https://publications.idiap.ch/attachments/reports/2006/vinciarelli-idiap-rr-06-75.pdf PUBLIC Idiap-RR-75-2006 2006 IDIAP Submitted for publication This work presents and compare two approaches for the semantic segmentation of broadcast news: the first is based on Social Network Analysis, the second is based on Poisson Stochastic Processes. Preliminary experiments address the problem of segmenting automatically different episodes of the same program into their two main sections. The experiments are performed over 27 hours of material: preliminary results are obtained by addressing the problem of splitting different episodes of the same program into two parts corresponding to a news bulletin and a talk-show respectively. The results show that the transition point between the two parts can be detected with an average error of around three minutes, i.e. roughly 5 percent of each episode duration.