The COST278 broadcast news segmentation and speaker clustering evaluation - overview, methodology, systems, results
Janez Žibert, France Mihelič, Jean‐Pierre Martens, Hugo Meinedo, João Paulo da Silva Neto, Laura Docío-Fernández, Carmén García Mateo, Petr David, Jindřich Žďánský, Matúš Pleva, Anton Čižmár, Andrej Žgank, Zdravko Kačič, Csaba Teleki, Klára Vicsi · 2005
This paper describes a large scale experiment in which eight research institutions have tested their audio partitioning and labeling algorithms on the same data, a multi-lingual database of news broadcasts, using the same evaluation tools and protocols. The experiments have provide more insight in the cross-lingual robustness of the methods and they have demonstrated that by further collaborating in thedomains of speaker change detection and speaker clustering it should be possible to achieve further technological progress in the near future.