In this paper we present different sources of information complementary to audio-visual (A/V) streams and propose their usage for enriching A/V data with semantic concepts in order to bridge the gap between low-level video analysis and high-level analysis. Our aim is to extract cross-media feature descriptors from semantically enriched and aligned resources so as to detect finer-grained events in video. We introduce an architecture for complementary resources analysis and discuss domain dependency aspects of this approach connected to our initial domain of soccer broadcasts.